Geo-Business GIS in The Dig PDF

fm JWPR043-Geo-Business October 19, 2007 12:4
GEO-BUSINESS: GIS IN THE

DIGITAL ORGANIZATION
Geo-Business GIS in the Digital Organization James B. Pick

Copyright © 2008 John Wiley & Sons, Inc.
i
This book is dedicated with thanks to my wife

Dr. Rosalyn Laudati and to the memory of my
father-in-law Hector Laudati.
ii
GEO-BUSINESS GIS IN THE

DIGITAL ORGANIZATION
JAMES B. PICK
University of Redlands
Redlands, California
JOHN WILEY & SONS, INC.
iii
This book is printed on acid-free paper.

∞
Copyright
C 2008 by John Wiley & Sons, Inc. All rights reserved
Published by John Wiley & Sons, Inc., Hoboken, New Jersey

Published simultaneously in Canada
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in

any form or by any means, electronic, mechanical, photocopying, recording, scanning, or
otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright
Act, without either the prior written permission of the Publisher, or authorization through
payment of the appropriate per-copy fee to the Copyright Clearance Center, 222 Rosewood
Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 646-8600, or on the web at
www.copyright.com. Requests to the Publisher for permission should be addressed to the
Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030,
(201) 748-6011, fax (201) 748-6008, or online at www.wiley.com/go/permissions.
Limit of Liability/Disclaimer of Warranty: While the publisher and the author have used their
best efforts in preparing this book, they make no representations or warranties with respect to
the accuracy or completeness of the contents of this book and specifically disclaim any
implied warranties of merchantability or fitness for a particular purpose. No warranty may be
created or extended by sales representatives or written sales materials. The advice and
strategies contained herein may not be suitable for your situation. You should consult with a
professional where appropriate. Neither the publisher nor the author shall be liable for any
loss of profit or any other commercial damages, including but not limited to special,
incidental, consequential, or other damages.
For general information about our other products and services please contact our Customer
Care Department within the United States at (800) 762-2974, outside the United States at
(317) 572-3993 or fax (317) 572-4002.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in
print may not be available in electronic books. For more information about Wiley products,
visit our web site at www.wiley.com.
Library of Congress Cataloging-in-Publication Data:
Pick, James B.
Geo-business GIS in the digital organization / James B. Pick.
p. cm.
“Published simultaneously in Canada.”
Includes index.
ISBN 978-0-471-72998-3 (cloth)
1. Management–Geographic information systems. 2. Business–Geographic information
systems. I. Title. II. Title: Geo-business geographic information systems in the digital
organization.
HD30.213.P53 2007
658 .05–dc22
2007013692
Printed in the United States of America
10 9 8 7 6 5 4 3 2 1
iv
CONTENTS
PREFACE xiii
ACKNOWLEDGMENTS xix
1. GIS IN THE DIGITAL ECONOMY 1

Business in the Digital Economy / 1
r The Rise in the Internet Platform for GIS / 2
r Spatial Data / 6
Case Study: Global Integrated Oil / 9
The Spatial Re-Evolution / 14
r Historical Antecedents / 14
r Coupling Technologies / 17
r Broadband as the Foundation for Web Services / 20
r Spatial Web Services / 21
r Open Source Software / 22
r Event-Driven Architecture / 23
The Spatial Industry / 24
Strategy, Competitive Advantage, and GIS / 31
Sperry Van Ness: Gaining Competitive Advantage
from GIS / 32
Chapter Summary / 34
References / 34
v
vi CONTENTS
2. INFORMATION SYSTEMS AND GIS 38

Introduction / 38
Overview of Information Systems / 40
GIS: Concepts and Definitions / 43
r GIS Location and Coordinates / 46
r Map Layers and their Relationship to
Attribute Tables / 48
r Spatial Analysis—What It Is and an Example / 49
Linkages to Other Disciplines / 51
Case Study: Baystate Health / 53
Understanding Spatial Data and Mapping / 54
r Digitizing / 55
r Scanning / 56
r Use of GPS / 56
r Satellite and Remote Sensing / 58
GIS Analysis / 58
r Spatial Analysis / 59
r Buffer Analysis / 61
r Map Overlay Functions / 61
r Proximity Analysis / 61
r Longitudinal Change Detection / 62
r Modeling and Forecasting / 62
r Statistical Analysis / 62
Example: Location Quotient Analysis for Siting / 62
National Spatial Data Infrastructures / 64
GeoWeb / 64
Future Trends in Spatial Technologies / 65
References / 66
3. SUPPORTING BUSINESS DECISIONS 70

Introduction / 70
GIS to Support Decision-Making in Organizations / 71
Case Study of Spatial Decision-Making in a Large
Personal/Corporate Bank / 79
r (1) Trade Area Analysis / 79
r (2) Market Demographics, Competitor Analysis, and
Customer Satisfaction / 80
CONTENTS vii
r (3) Branch-Related Campaigns / 80

r (4) LPCB Securities / 80
r (5) Mapping Support for Executives on Strategy / 80
Spatial Decision Support Systems / 81
r SDSS and Visualization / 82
r Prevalence of SDSS / 84
Spatial Business Intelligence Systems / 90
r WebFocus from Information Builders / 91
r Map Intelligence / 92
Case Study of Large Insurance Company / 93
References / 95
4. ENTERPRISE APPLICATIONS 99
Introduction / 99
GIS and Enterprise Resource Planning / 102
Case Study of ERP at Pidpa, Belgian Water Utility / 107
GIS and Customer Relationship Management / 110
r Site Evaluation Models / 114
r Distribution of Resources / 114
Chico’s Case Study: GIS and CRM in a Customer-Centric
Fashion Company / 115
GIS and Supply Chain Management / 118
Spatial Data Warehouses and Data Mining / 122
Case Study: Nesa, Danish Electrical Utility / 124
References / 129
5. CUSTOMER-FACING GIS: WEB, E-COMMERCE, AND

MOBILE SOLUTIONS 133
Introduction / 133
Architectural Approaches to Spatial Technologies / 134
r Client-Server / 134
r Web Services / 138
Consumer Web Services / 142
r Trulia.com / 145
r DataPlace.org / 145
viii CONTENTS
Mobile Services / 147

r DS Waters / 149
r Seaspan International Ltd. / 152
Case Study: MotionBased Technologies / 154
E-Commerce / 155
r MasterCard / 157
r Zillow.com / 158
r Market Efficiency / 159
Standardization / 161
r Grid Computing / 165
References / 167
6. SPATIAL SYSTEMS DEVELOPMENT 169

Introduction / 169
Systems Development for GIS / 170
Phases in Systems Development for GIS / 174
r The Planning Phase / 174
r The Analysis Phase / 176
r Design / 178
r The Implementation Phase / 184
r The Maintenance Phase / 185
Case Study Findings on Systems Development / 185
r Was Systems Development Done in-House or Mostly
Outsourced? / 186
r If in-House, were Formal Phases Taken, or were
Information Steps Followed? / 187
r What was the Composition of the Systems Development
Project Team? / 187
r Was the Systems Development Approach Traditional or
Object-Oriented? / 187
r Were there Differences by the Size of Company in the
Approach Taken? / 187
Case Study: Engineering Systems / 190
References / 194
7. THE VALUE OF INVESTING IN GIS 195

GIS and Geographic Information—Their Value / 195
Setting the Context for Cost-Benefit Analysis / 196
CONTENTS ix
The Costs and Benefits of GIS / 199

r Costs / 199
r Benefits / 200
Prevalence of Costs and Benefits in Case
Study Companies / 202
r Unit of Analysis / 207
The Substeps in Cost-Benefit Analysis from Systems Design
Perspective / 208
Comparison of Costs and Benefits Over Time / 210
The Method of Discounted Cash Flow / 211
Comparison with a Criterion / 211
Intangible Costs and Benefits / 212
The Feasibility Decision / 213
Stakeholders and Externalities / 213
Case Study of GIS Costs and Benefits: Sears Roebuck / 214
The Productivity Paradox for GIS / 218
Case Study: Norwich Union Unexpected
Benefits of GIS / 220
Acknowledgments / 221
References / 222
8. MANAGING SPATIAL DATA 227

Introduction / 227
Introduction to Managing Spatial Data / 228
GIS, Spatial Data, and Relational Databases / 231
r Spatial Relational Data Model / 232
r Spatial Object-Oriented Data Model / 235
r Spatial Object-Relational Data Model / 238
r Oracle Spatial 10g / 239
r ESRI’s Geodatabases / 243
Case Study of Enmax / 246
GIS and Data Warehouses / 248
r Definition / 248
r Historical Background / 249
r Fundamentals of the Data Warehouse / 249
r Key Applications of Spatial Data Warehouses / 254
r Future Directions in Data Warehouses / 255
Managing Data Quality / 256
x CONTENTS
Acknowledgment / 258
References / 258
9. ORGANIZATIONAL AND INDUSTRY ASPECTS 261

GIS and the Organization / 262
GIS in Industry Areas / 264
Utilities / 267
Case Study of Southern Company / 278
Transportation / 280
Retail / 286
References / 293
10. INDUSTRIES AND OUTSOURCING 295

Introduction / 295
Insurance / 296
Swiss Reinsurance Company / 298
Marketing/Geodemographics / 302
Banking and Finance / 311
Case Study of Large Credit Bank / 312
Agriculture / 314
Health Care / 317
Spatial Technologies and Outsourcing / 319
Kaiser Permanente / 321
References / 324
11. ETHICAL, LEGAL, AND SECURITY ISSUES OF SPATIAL

TECHNOLOGIES 327
Introduction / 327
Ethical Issues / 328
Privacy / 336
Legal Issues / 339
Case Study of Legal Aspects of Surveillance: Kyllo v. United
States / 346
Crime and Security / 347
References / 351
CONTENTS xi
12. GIS AND BUSINESS STRATEGY 355

Strategic Vision for GIS and Establishing a Strategic Plan / 356
Theories of IT Alignment, Internet Strategy and Situational
Analysis / 356
r IT Alignment / 357
r Internet Strategy / 359
r Situational Analysis / 360
Case Study: Norwich Union / 361
An Evolutionary Framework for Strategic GIS / 364
Findings of Case Study Analysis / 367
Case Findings for High Strategic Firms / 367
r Global Integrated Oil / 367
r MotionBased Technologies / 367
r Rand McNally / 368
r Sears Roebuck / 368
r Southern Company / 368
r Arizona Republic / 369
r Chico’s / 369
r Lamar Advertising Company / 370
r Prudential Preferred Realty / 370
r MapGistics / 370
Case Findings for Medium Strategic Firms / 371
r Kaiser Permanente / 371
r Large Insurance Company / 371
r URS / 371
r Sperry Van Ness / 372
r Western Exterminator / 372
Case Findings for Low Strategic Firms / 372
r Large Personal/Corporate Bank / 372
r Large Credit Bank / 372
r Norwich Union / 373
r Baystate Health / 373
r Engineering Systems / 373
Summary of Findings / 374
Practical Advice to Managers on Strategies for GIS and Spatial
Technologies / 377
r Matching the GIS Strategic Plan with the Corporate
Strategic Plan and the Corporate IT Plan / 377
Case Study: Rand McNally / 378
xii CONTENTS
The Future and Spatial Strategies in Enterprises / 381

References / 383
INDEX 387
PREFACE
Geographic Information Systems (GIS) support business decision-making by

including a spatial component in the decision process. With GIS, businesses
plan and make judgments based on more than the attributes of business
entities, but on their spatial properties as well. Although GIS gained its early
adoption mostly in government, today it is expanding rapidly in business.
New corporate uses and applications are being discovered and developed.
The overall goal of the book is to provide a comprehensive coverage of
GIS applied to and utilized by businesses and organizations. Unlike textbooks
with a mainly geographical perspective, this book sets GIS in the context
of the concepts and practices of information systems and other business
sub-disciplines such as marketing, logistics, business ethics, management,
and strategy. The real world is brought in through case studies, examples,
explanations of technologies, and up-to-date references. For more advanced
readers, the book adds to and updates their detailed knowledge of the latest
trends and developments.
The book’s mission is to thoroughly educate the reader on the principles
and capabilities of GIS and to make managers and their businesses smarter,
more efficient, productive, decisive, and ethically aware.
This book is written to fill in a gap in the knowledge of GIS and spatial
technologies. Although business uses and applications have been included in
books, they have not centered on well-organized frameworks of business, and
information systems (IS).
When GIS is practically applied in the real world, it has a strong manage-
ment component. Technical knowledge or geographic experience alone will
not lead to successful GIS implementations. Failures have occurred most of-
ten because people and organizations did not communicate, leadership did not
xiii
xiv PREFACE
support or take interest in a spatial project, customers were not consulted, or

the project was not scoped well and benefits assessed. These are not technical
or geographic but management issues.
Likewise, GIS depends on information systems. The IS that underpins a
GIS must meet the normal IS success criteria of user-responsiveness, usability,
fit to the task, reliability, and maintainability, A GIS implementation needs
to be conceptualized, designed, and built according to the best principles and
practices from IS. Data need to be organized and well managed, which largely
derives from IS. Controls for a GIS closely resemble those of information
systems. As GIS expands to enterprise systems, the web, and mobile devices,
much depends on the information technology.
Another perspective on the dependence on IS is reflected in the location of
GIS units in organizational structures. For mid- and large-sized companies, the
GIS unit is commonly in a separate small unit closely aligned and interacting
with an older and larger IS unit. Sometimes GIS is brought right into the GIS
unit. Either way, they are closely linked.
One of the challenges in the geobusiness area of study is paucity of aca-
demic research. For instance, in the decision-making literature, there were
only a handful of research journal articles over the past decade on GIS
decision-making (Huerta et al., 2005). Because the academic literature on
GIS in business is scattered in a variety of types of journals, it takes some
detective work to find it. One reason there is little literature relates to the
proprietary nature of many spatial systems in business. If a major company is
gaining competitive advantage from GIS, it might wish not to make it known
and not open up that information to academic investigators.
Hence, the project needed to delve more into the recent professional and
trade literature. Here, astute business observers and experts provided rich
practical information about what is happening in geobusiness. Another source
of such information were several annual conferences especially the Geospatial
Information and Technology Association (GITA) Conference, the Location
Intelligence Conference, the ESRI Business GeoSummit, and the Annual
ESRI International User Conference.
The book has a total of 26 case studies. This includes twenty research case
studies that were analyzed using rigorous case study methodology. Sometimes
the term "research case" is used, which refers to a case in this group of 20. In
addition, six cases were drawn from secondary references, which are included
because they particularly reinforce points in the book. The cases and their
chapter coverage are given in the table below.
For each research case, the corporate manager of GIS was interviewed
for an average of two hours, following a protocol of questions. The inter-
view protocol can be obtained from the author. Permission to record was
largely granted and the interviews were transcribed. Other supporting docu-
ments were requested from the interviewees. The results were analyzed using
standard case study methods (Yin, 1994). The findings are included in the
book in several ways. Fourteen of the twenty research cases have narrative
PREFACE xv
Case Study Firms in GeoBusiness

Name or description Chapter Size of Firm
Research Case Studies (based on primary case study interviews)
Global Integrated Oil (GIO) 1 large
Sperry Van Ness 1 medium
Bay State Health 2 medium
Large Corporate/Personal Bank (LCPB) 3 large
Large Insurance Co.(LIC) 3 large
Chico’s 4 medium
Motion-Based Technologies 5 small
Engineering Systems 6 small
Sears Roebuck 7 large
Southern Company 9 large
Kaiser Permanente 10 large
Large Credit Bank (LCB) 10 large
Norwich Union 7, 12 large
Rand McNally 12 large
Arizona Republic 12 - summary medium
Lamar Advertising Co. 12 - summary medium
MapGistics 12 - summary small
Prudential Preferred Realty 12 - summary medium
URS Inc. 12 - summary large
Western Exterminator 12 - summary medium
Additional Textbook Case Studies (based on secondary sources)
Pidpa 4 medium
Nesa 4 medium
DS Waters 5 small
Seaspan International Ltd 5 medium
Enmax 8 medium
Kyllo v United States 11 NA
NA = not applicable
case-study sections in the book. In Chapter 12, short profiles appear of all
twenty research case firms, including the six that do not have separate chapter
sections. All twenty or subsets of them are included in analyses and compar-
isons throughout the book.
For each research case, the firm was given the choice to remain anonymous,
and four elected to do so. The firms allowing their names to be included are
listed in the table, along with the four firms that elected anonymity, which were
given the pseudo names of Large Credit Bank, Global Integrated Oil, Large
Insurance Company, and Large Personal/Corporate Bank. In the table, their
initials appear, which are sometimes used to refer to them. The information
and knowledge flowing from the interviews was impressive and enriches the
research cases throughout the book.
xvi PREFACE
In addition to the primary research case studies, six other case studies are
included in the book derived from secondary sources because they particularly
illustrate chapter themes. They are Pidpa and Nesa, Belgian and Danish utility
firms; DS Waters in Atlanta; and the Canadian firms Seaspan International
Ltd and Enmax. Four of them are non-U.S. which gives more international
flavor to the book overall, and reflects that spatial technologies are in use
worldwide. Also a legal case study is included in Chapter 11 of Kyllo v.
United States.
GIS and spatial technologies are currently moving towards the web. This
reflects a broader trend of information systems generally towards the internet
platform. Spatial capability is becoming prevalent in mobile devices, starting
with GPS-enabled cell phones required in many countries. The data bank of
spatial information is growing exponentially through collection by multiple
types of sensors, satellites, RFID readers, and GPS-enabled devices. Data
storage technologies and designs have advanced into spatial object-oriented
and spatial database paradigms. Large-scale consumer web mapping appeared
with Google Maps in 2005. The appearance of widespread RFID in supply
chains in the U.S. was marked by Wal-Mart’s mandate to its largest suppliers
to make the shift in 2005. The book is fortunate to appear while these rapid
changes are underway and can document and explain them.
In covering GIS for business, study of management and technology needs
to be balanced with learning about the ethical, legal, and security dimensions.
For instance, there are ethical and privacy issues related to the dissemination
and use of sensor-based information. Chapter 11 examines what these issues
are and how some of them are being resolved while other await future reso-
lution. The treatment is not prescriptive, but rather informative—to expand
awareness for the reader’s benefit in viewing these issues and possibly having
to formulate stances and make personal choices related to them.
The book is intended for several audiences. Managers in industry and
organizations untutored in GIS and spatial technologies can benefit by un-
derstanding the business fundamentals of these phenomena and the potential
to implement them in their companies. For skilled GIS managers, the book
fills in gaps, provides some new ideas, and allows them to benchmark and
compare with other companies. They can learn from seeing the successes
and mistakes of others. The book stresses how well-established business and
IS concepts can be applied to spatial problem-solving, and at the same time
what is different and special about spatial systems. The book does not devote
much space to geographical principles, which are well covered in many other
textbooks. A business manager needs to be knowledgeable but not necessarily
an expert in geography and GIS to be successful with spatial applications and
projects, but he or she does need to be expert in management principles and
practice in order to succeed. He/she can gain access to GIS technical expertise
in-house or through external sourcing.
Another audience is academic. In teaching, the book is suitable for upper
division and graduate courses in business, public administration, planning,
PREFACE xvii
environmental management, applied geography, and related disciplines. The

book can contribute to improving the preparation of people entering a growing
geospatial workforce (Marble, 2006).
The book is also intended for academic researchers in business, manage-
ment, IS, planning, public administration, and applied geography disciplines
who need an integrated and broader context in which to set their research.
The book refers to many academic studies across a variety of subfields. The
case study analysis in the book can be helpful to some on particular research.
It is the intent and hope that GeoBusiness will provide currency of ideas,
new insights, practical benefits, and motivation to people already working in,
or with interest in GIS and spatial applications in business.
REFERENCES
Huerta, Esperanza, Celene Navarrete, and Terry Ryan. 2005. "GIS and Decision-
Making in Business: A Literature Review." In Pick, James B. (Ed.), Geographic
Systems in Business, Hershey, PA: Idea Group Publishing, 20–35.
Marble, Duane F. 2006. "Who Are We: Defining the Geospatial Workforce." Geospa-
tial Solutions, May, 14–21.
Yin, R.K. 1994. Case Study Research: Design and Methods, Second Edition.
Thousand Oaks, CA: SAGE Publications.
ACKNOWLEDGMENTS
This project would not have been possible without the support, cooperation,
and collaboration of many individuals.
Thanks to several people who provided early ideas at the inception and
support: Julian Ray, formerly assistant professor at University of Redlands;
John Stager, doctoral student at Claremont Graduate University; and Paul
Gray, professor emeritus at Claremont Graduate University. Thanks also to
Joe Francica, Peter Keenan, Xavier Lopez, and Bruce Ralston, who reviewed
the proposed book content.
Thanks to chapter reviewers Rob Burke, Peter Keenan, Ray Papp, Monica
Perry, Dave Petrie, Keith Roberts, Namchul Shin, Jeffery Smith, John Stager,
Larry West, Vijay Sugumaran, and Nanda Viswanathan for their insightful
critical comments, suggestions, and advice. Any errors that remain are the
sole responsibility of the author.
Appreciation is expressed to Hindupur Ramakrisna for discussions on
spatial decision support systems. The students in the fall 2006 MBA
Connections-B course on GIS and faculty colleague Kimberly Cass were
helpful in pre-testing several chapters of the draft manuscript. Their ideas
and feedback on the material were helpful and most appreciated. Apprecia-
tion is expressed to the libraries of University of Redlands and University of
California Irvine for the high quality and extent of their resources.
The School of Business at University of Redlands has been supportive in
a number of ways. School research funding was helpful and is acknowledged
for the final stages of the project. Faculty feedback is appreciated from sev-
eral presentations related to the book in the school’s research series. Thanks
to the school’s faculty support office of Joanie James, Shari Audelo, Susan
Griffin, and Erin Ross, as well as undergraduate student assistants, who were
xix
xx ACKNOWLEDGMENTS
consistently helpful. Special thanks to MBA student Matt Wick for his out-
standing assistance, support, and ideas on organizing digital materials, ob-
taining permissions, and discussing book issues and to undergraduate student
Matt Rogers for strong and energetic support at the end of the project.
Thanks to the GIS managers and leaders at twenty case study companies.
Appreciation is expressed to Jill Boulton, Danny Childs, Amy Claar-Pressley,
Joe DeVoy, Kurt Gunther, Steve Jones, Mike Lawton, Chris McKeever, Joel
Minster, Jason Murray, Swapan Nag, Ken Pitts, Sheryl Poss, Tom Sanchez,
Steve Schonhaut, Ric Skinner, Jay Visnansky, Clark Weber, as well as to
individuals in the four companies that elected to remain anonymous. Special
thanks to Swapan Nag, for some in-depth discussions subsequent to the
interview and to Joe DeVoy who after the interview provided more ideas and
made his office open to a field visit.
At ESRI Inc. thanks to Jack Dangermond for general support and ideas
and to Steve Benner, Ann Bossard, Bill Davenaill, Roxanne Cox-Drake,
Sean Fitzpatrick, Randy Frantz, Russ Johnson, Jim Herries, Lisa Horn, Bill
Meehan, Alex Miller, Lew Nelson, Mike Phoenix, Simon Thompson, and
Geoff Wade for their ideas and support.
The author had a sabbatical leave at University of California Irvine in
summer and fall of 2005, which provided a quiet office and opportunities
to discuss and present book ideas, moving the project forward significantly.
Appreciation is given to Frank Bean in School of Social Sciences for arranging
the sabbatical at UCI and encouragement and to Tony Soeller in Network and
Academic Computing Services for helping to formulate some book ideas.
Thanks to Jim Harper of John Wiley and Sons for his insights on GIS and
for editorial advice.
The author wishes to acknowledge all these people and many others not
mentioned who provided assistance and backing.
Finally the author expresses special thanks to his wife, Dr. Rosalyn Laudati,
for her great understanding, patience, and support during the many long hours
of the project.
c01 JWPR043-Geo-Business October 18, 2007 13:20 Char Count=
CHAPTER 1
GIS IN THE DIGITAL ECONOMY
BUSINESS IN THE DIGITAL ECONOMY
Spatial technologies and GIS are impacting the productivity of business and
economies. As information technologies have become more pervasive, inter-
active, mobile, internet-based, and diffused throughout the enterprise, likewise
spatial technologies have done so. Information technologies became prevalent
for mainframes in the late 1950s and early 1960s, while GIS only appeared
commercially in the late 1960s and became widespread in government in
the late 1970s. This lag means that information technologies became well
established in organizations about fifteen years earlier than GIS. In the busi-
ness world, the lag is greater because GIS was first adopted by governments,
remained largely a public-sector feature for two decades, and only became
widespread in businesses in the 1990s.
The reasons that GIS has not caught on until recently in the business sector
include its high cost and lack of perceived benefits. Spatial datasets are larger
than corresponding non-spatial ones. Often spatial data analysis requires more
processing power than similar non-spatial analysis. Thus, computing capac-
ity parameters took longer to provide enough speed and power to adequately
support spatial analysis and its applications. Second, business people have
had less overall knowledge of spatial principles and applications than their
government counterparts. Until recently, exposure to spatial software in in-
dustry was limited and fairly expensive. Training is sometimes difficult for
the average business person to obtain. Software training is provided by GIS
software vendor firms and many community colleges. An example is the GIS
1
2 GIS IN THE DIGITAL ECONOMY
Education at City College of San Franciso (CCSF, 2007). Universities have

not until recently emphasized the business side of spatial technologies. GIS is
a prevalent tool in schools of planning and public administration, but is only
beginning to take hold in business schools (Pick, 2004). Generally, universi-
ties have not understood what is needed to educate and train the geospatial
workforce (Marble, 2006). Third, business top management leadership in the
1990s infrequently recognized the importance of spatial dimensions. Today
more business leaders do recognize it. As seen in cases in the book, some
top leaders in companies that heavily utilize GIS/Spatial technologies do
recognize its significance strategically, while others do not.
The Rise in the Internet Platform for GIS

Another set of trends that has stimulated spatial and GIS applications is the
rise of the internet, web, and e-commerce applications. The following recap
of the development of the internet, underscores how recently web-based GIS
has been possible and how it serves as a driver for GIS.
Although the application of the internet started in 1969 at the U.S. Depart-
ment of Defense Advanced Research Projects Agency (DARPA), the main
protocol of TCP/IP was not developed until 1974 and only came into research
and academic use at leading centers in the late 1970s. TCP/IP (transmis-
sion control protocol/internet protocol) is the protocol that controls internet
communications between computers. The management of the internet shifted
from Department of Defense (DOD) to the National Science Foundation in
1987 and to international commercial organizations in 1995. At that point, the
backbone internet traffic was taken over by commercial telecommunications
carriers worldwide such as MCI Inc., AT&T, Sprint, and Nippon Telephone
Company. As seen in Table 1.1, the number of internet hosts worldwide in-
creased steadily from the mid 1980s to 2005 (The Internet Society, 2006). An
internet host is any computer system connected to the internet from a full-
or part-time, direct or dial-up connection (more specifically, a host is any
computer with an IP address). Since internet hosts only exceeded 100 million
in 2000, this table underscores why widespread GIS and spatial applications
on the internet are recent.
The World Wide Web (WWW) was originated in 1989 at the European
Laboratory for Particle Physics by Tim Berners-Lee. It allowed a user to jump
around the internet by links attached to text and imagery. The first browser,
Mosaic, was invented in 1993 at the National Center for Supercomputing
Applications at University of Illinois. This browser and others that followed
such as Microsoft Explorer (1995) led to standard and user-friendly interfaces
that made the web navigation much easier.
Businesses had been able to use the internet to conduct some business
transactions in the 1980s through Electronic Data Interchange (EDI), but the
control of the interchanges was not user-friendly, so it was mostly special-
ists in IT departments who developed, operated, and utilized EDI. With the
advent of the WWW, a new form of business exchange became available,
BUSINESS IN THE DIGITAL ECONOMY 3
TABLE 1.1 Number of Internet Hosts, 1984–2005
Year (January) Number of Internet Hosts
1984 1,000
1991 376,000
1992 727,000
1993 1,300,000
1994 2,200,000
1995 5,000,000
1996 9,400,000
1997 16,000,000
1998 29,000,000
1999 43,000,000
2000 72,000,000
2001 109,000,000
2002 147,000,000
2003 171,000,000
2004 285,000,000
2005 318,000,000
Source: Internet Systems Consortium, Inc. (http://www.isc.org/).
electronic commerce. Electronic commerce is the conduct of commercial

transactions on a widespread basis through internet-based exchanges, mostly
WWW-based. It grew in the early to mid 1990s and today is estimated to
underlie about 5 percent of the U.S. economy, a proportion that is grow-
ing. E-commerce can be business-to-consumer (B2C) or business-to-business
(B2B). Business-to-consumer e-commerce consists of business transactions
and exchanges between a customer and the business; for instance, if a con-
sumer on the web purchases a laptop computer from Dell, or books from
Amazon. Business-to-business involves transactions and exchanges between
two or more businesses. An example is a B2B website for chemical compa-
nies to sell and purchase chemical products with other chemical firms. Often
B2B applications provide the basis for market transactions.
The expansion of e-commerce in the United States has been rapid and
continues. For instance, e-commerce as a percentage of U.S. retail sales
grew steadily from 0.91 percent in 2000 to 2.37 percent for year 2005 (U.S.
Census, 2006). E-commerce trends relate to GIS applications, since spatial
interfaces are beginning to be involved. This can be through B2C user inter-
faces with spatial features, marketing of e-commerce, and delivery of goods.
For instance, many Google Earth mash-ups, i.e. third party products based on
Google code, allow customers to search for business products and services,
and conduct online business transactions. The Sperry Van Ness case at the
end of the chapter supports spatial features on the web so brokers and their
customers can more quickly research, market, and close commercial real
estate sales transactions.
Since spatial systems tend to lag information systems (IS) somewhat, web-
based spatial applications have only appeared heavily since 2000. Today the
shift in spatial applications is steadily towards the web and internet (Sonnen
et al., 2004; Sonnen, 2006; Daratech, 2004). Users find the internet platform
appealing, easy, and flexible between devices. However, the web protocols
and the designs, servers, and software to support these spatial applications
are still in development. For instance, the leading spatial web protocols such
as GML (Geography Markup Language),WFS (Web Feature Services), and
WMS (Web Map Service) are available through a leading standards body, the
Open Geospatial Consortium (OGC), but are not yet fully accepted industry
standards.
E-commerce applications with spatial components became evident in the
Dot.Com boom of the late 1990s with the advent of real estate, transport
routing, and other web-based services with map features. The e-commerce
with spatial features is growing and particularly relates to B2C in such sectors
as real estate, retail, tourism, transportation, and distribution. These spatial
advances became evident in 2005 to larger audiences of hundreds of millions
of internet users through the milestone advent of Google Earth, Google Map,
Microsoft Virtual Earth, and Yahoo Map, and smaller “mash-up” applications.
For example, a person ordering a pizza online compares pizza parlors based
on their location and the web links describing them, and then ordering online
at Pizza Hut (see Figure 1.1).
The breakthroughs in 2005 also swept through the GIS industry and in-
fluenced GIS software companies to undertake new strategies of web-based
Figure 1.1 Spatial E-Commerce: Ordering a Pizza Online from Google Map Dis-
play. Source: Google MapsTM mapping service/NAVTEQ 2007
applications that are broadening and changing spatial applications for varied
businesses, large and small, across many vertical sectors.
This trend towards the spatial-web is a recurring theme in this book, in the
chapters that explain different aspects of GIS in business, and also in many
case studies. The Sperry Van Ness case demonstrates how a medium-sized
firm can make wise choices of internet platforms and software, not try to do
too much, and be highly successful today in the spatial-web space.
Also in the 1990s, large-scale, enterprise-wide software applications be-
came more prevalent (Gray, 2006). These include Enterprise Resources Plan-
ning systems (ERP), Customer Relationship Management (CRM), and Supply
Chain Management (SCM). The difference from the functional systems avail-
able earlier is that these systems apply widely across all locations and many if
not all divisions of a business. ERP supports integrated transaction processing
systems across wide functional areas including accounting, finance, market-
ing, sales, production, human resources, and inventory. CRM systems support
managing the long-term relationships with the company’s customers, from
initiation, through building and development and growing breadth of rela-
tionships, to transfer, termination, or upgrading of relationships (Gray, 2006).
Supply Chain Management (SCM) monitors, manages, and projects compa-
nywide flow of raw materials, components, and finished products throughout
the manufacturing, distribution, and delivery processes (Gray, 2006). It not
only tracks physical items, but also the associated flows of information. If the
materials, components, and products moving through the supply chain are
geo-referenced at many points in the process, then the supply chain can be
better understood, tightened up, optimized, and made more predictable than
without geo-referencing.
Geo-referencing refers to adding X-Y (longitude-latitude) fields to an ex-
isting data record. Although less prevalent, this might be in 3-D, i.e. X-Y-Z
(longitude-latitude-elevation). 3-D geo-referencing can be used for terrain el-
evation modeling and other applications. Roughly 80 percent of business data
has the potential to be geo-referenced, i.e. have a spatial location attached to
it (Bossler, 2002).
Because of its greater prevalence of web-based architecture, GIS is becom-
ing more strategic in its applications, often extending across the enterprise.
It is no longer restrained to traditional, 1990s departments that maintained
compartmentalized datasets not accessible to the outside. Rather, following a
long-term IT trend, spatial systems are beginning to be incorporated as a key
part of enterprise-wide business applications such as ERP, CRM, and Supply
Chain. The process of its movement into these domains has been inconsistent,
bumpy, and sometimes resisted by management and users. Another constraint
has been how to seamlessly bridge between the large enterprise business ap-
plications, which were originally developed without spatial modules, and the
complex GIS software packages. However, vendors of both types of soft-
ware, such as SAP and ESRI Inc., see the opportunity and are working on
smoother and more efficient connecting interfaces. In medium and large firms,
the enterprise applications were often implemented by IT specialist groups

that were not grounded in spatial principles and often isolated from usually
small GIS departments, creating other obstacles that have slowed the pace
of development. Of the book’s twenty research case studies, few yet have
implemented GIS integrated with enterprise applications, and none fully.
Spatial Data
When spatial and GIS applications take place in business, whether traditional
client-server systems or enterprise-web-based ones, they are based on business
data. High quality spatial data (boundaries and attributes) are critical because
they support any spatial application and serve as the foundation for analysis,
modeling, and decision-making. Spatial data consists of two parts:
r the digital map boundaries, which constitute the map layer or map cov-
erage
r data associated with the map layer, which are commonly alphanumeric
but can also be video images, audio, and other forms. The non-spatial
associated data are referred to as attribute data, and are considered more
in Chapter 2.
The role of the two types of data can be seen in the simplified Generic Design
of a GIS, shown in Figure 1.2. The digital boundary data on the bottom right
are input from the internet, global positioning systems (GPS), satellites, and
internal sources, and provide the digital boundaries for each map layer. The
attribute data on the bottom left are input from other internal and external
sources, and provide the non-spatial attributes associated with spatial features.
For example, the diagram’s middle boundary layer shows three roads. The
non-spatial features of each road (width, materials, date of construction, date
of most recent maintenance) are stored in an attribute table associated with
the map layer (shown on the left). The model of GIS is explained in detail
in Chapter 2, while spatial data are emphasized in Chapter 8. The Model
also has analysis and modeling functions, which depend on the boundary and
attribute data. This includes functions for statistics, simulation, forecasting,
and spatial analysis. Finally the outputs at the top are the processing results
of the GIS—maps, graphical displays, tables, and other information that is
provided to the user for decision-making.
Important aspects of spatial data are its cost, ownership, security, privacy,
data quality, and currency/updating. The need to assure all this makes data
acquisition often costly and time-consuming for the following reasons:
r A well-known estimate is that 80 percent of the development costs of a

GIS project are data expense (Bossler, 2002). The reasons are that spatial
data are often voluminous, come from varied data sources, may have
Output
Analysis and
Modeling
Attribute Tables Map

Associated with Layers
Each Boundary
Layer
Attribute Data-
Input Digital Boundary
Data-Input
Figure 1.2 Generic Design of a GIS, Showing Attribute Data and Digital Boundary
Data
quality issues, and they may not be compatible with each other or with
the GIS software. The latter issue stems from the GIS industry’s need to
update and replace standards, as a consequence of business and technical
change, a topic addressed in Chapter 5.
r Data quality is also a problem. Not only are there the usual problems
with the attribute data, but in addition serious challenges exist with the
accuracy of digital boundary files (Meeks and Dasgupta, 2005). Some of
the digital boundary data were originally collected before GPS (Global
Positionig Systems) and satellites provided precise locations. A promi-
nent example of this are the base boundary files (TIGER) of the U.S.
Census, which have been updated to GPS-accuracy for the 2010 Census.
Even with GPS and satellites, errors can occur in the processing and
management of the data, as well as in correctly identifying and labeling
location. If the user needs exact locations, such as for pipelines or utility
lines, then there is a need to do thorough checking for data accuracy. If
necessary, accuracy assessments with GPS against physical features in
the field may be necessary at extra time and expense to ensure the data
quality. Accuracy is also needed to assure the precise connectivity of
pipelines with each other.
r The data cleaning, checking, and modifications can exceed ordinary data
quality assurances because both the attribute and spatial data need to
be individually checked and often modified, but also they need to be
compatible with each other and with the GIS software.
r A large proportion of business-generated data are proprietary. This means
that they are either not released by a firm or only to trusted or partner
firms and often at cost. The firm has legal ownership of the data, so in the
event of disputes over ownership or data theft, threats or even lawsuits
can occur.
r Coordinate systems in boundary layers may be mismatched. Modern
software tools can help in achieving matching, but good technical under-
standing is often needed.
The scope of spatial and associated attribute data is vast and covers the
range of business activities in multiple industries. Some of the lead industries
that receive attention in this book are utilities, transportation, petroleum,
insurance, banking, real estate, retail, environmental consulting, and health
care. For just these, the scope of spatial data includes:
r Utilities. Attribute and boundary data on economic development, demo-

graphic patterns, vegetation, rates, loss evaluation, environmental con-
tamination, maintenance, work-order status, weather, safety, crew and
vehicle tracking, outages, transmission lines, land use, metering, labor,
materials, customers, and accounting costs.
r Transportation. Attribute and boundary data on transport routing, con-
gestion patterns, volume of traffic, vehicle types, loads, maintenance,
fuel types, customers, and accounting costs.
r Petroleum. Data on geology, drilling, exploration, land use, pipelines,
refineries, consumer markets, retail locations, customers, demographic
patterns, political patterns, and assets.
r Insurance. Data on land use, property damage, vegetation, weather, pre-
mium pricing, environmental risks, tax jurisdictions, customers, com-
petitors, natural catastrophe pricing, event loss estimation, population,
injuries and deaths, homeless and missing persons, peril zones, and port-
folio analysis.
r Banking. Data on branch and ATM locations, market areas, spatial diver-
sification of investments, and individual and business customer profiles
by location.
r Real estate. Data on property locations, layout of land and properties,
pricing history, markets—buyers and sellers, taxes, mortgage rates, and
customers.
r Retail. Data on inventory, supply chain, sales, store sites, customers,
transportation networks, demographic profiles and trends, competitors,
and purchase patterns.
CASE STUDY: GLOBAL INTEGRATED OIL 9
r Environmental consulting. Data on environmental contamination, envi-

ronmental treatments, management of clean-ups, and customer locations.
r Health care. Data on locations and attributes of doctors, other health-care
workers, patients, diseases, treatments, facilities, service and regulatory
boundaries, transportation, inventory, scheduling, and costs.
These and other data sources are considered in Chapters 9 and 10. Much of the
vast reach of business spatial data has existed for many decades, pre-dating
the advent of widespread business spatial uses in the 1990s. Since some of the
data was not originally accurately geo-referenced, data collection procedures
need to be modified so that old locations can now be brought up to the latest
precision. For instance, a New York durable goods retailer with thousands
of daily deliveries can record and check the X-Y locations of a customer
by recording it from the GPS device on the delivery truck, during a home-
delivery event. Geo-referenced records have both spatial and non-spatial uses.
For example, some companies have geo-referenced their inventory. This can
be done through older methods such as adding the location as items are
received manually at facilities, or through automatic recording by RFID (radio
frequency identification) devices. Knowing precisely where each item of
inventory is located helps improve efficiency and timing. Geographically
referenced inventory may or may not utilize the spatial referencing, depending
on the business process. For example, a chain of grocery stores can calculate,
by store, the monthly dollar value of its three current best-selling food items
and map those values by store. The map utilizes the spatial location i.e. the
point location of each store. Another analysis can do a line graph of the total
corporate monthly sales for the three current best-selling items over the past
ten years. This line graph is useful for corporate market planning but has no
spatial-referencing. This point becomes important with enterprise business
systems, which up to now have had minor spatial aspects. For example in the
ERP for one large business, over 98 percent of analysis is non-spatial and
only 2 percent is spatial. Hence, proportion of the non-spatial uses must be
considered in recognizing the perceived importance of spatial data and the
priorities given by managers and other stakeholders in utilizing it.
CASE STUDY: GLOBAL INTEGRATED OIL
Global Integrated Oil (GIO) is a world giant, with over 50,000 employees,
more than $100 billion in revenues, and business conducted in 180 nations.
Spatial technologies are applied enterprise-wide, including in exploration,
leasing, transportation and storage, environmental, refining, pipelines, mar-
keting, distribution, supply chain, and business planning. These areas can
be better conceptualized with a picture of its Energy Value Chain, shown in
Figure 1.3, from exploration to production through to distribution and mar-
keting to customers.
These steps are supported by a mixture of technologies that includes infor-
mation systems, e-commerce, mobile technologies, and GIS.
Figure 1.3 GIO’s Energy Value Chain
Some notable examples of GIO’s spatial activities are the following:

Upstream GIS uses are the heaviest. In exploration and development, geol-
ogists analyze the subsurface using 3-D modeling software, mostly purchased
from outside vendors. This is a competitive aspect, since the global petroleum
firms compete strongly on strategic knowledge of where petroleum reserves
are located. In determining where to explore, the upstream GIS groups model
the global petroleum basin formation from 500 million years ago to present.
For instance, Figure 1.4 shows the location of subsurface oil and deposits,
Cretaceous
95 MYBP
Figure 1.4 Geologic Model of Oil and Gas Deposits in the Cretaceous Period
as well as certain related fossil deposits in the Cretaceous Period that oc-
curred earlier than 95 million years ago. The darker the shading, the larger
the deposit.
Global discoveries by GIO over the past ten years of oil and natural gas
are mapped, along with locations of dry holes and tar sands. This helps
to strategically appraise the size and locations of exploration projects. The
benefits are knowing not only where to explore, but also which zones to avoid.
Once exploration has commenced, maps of seismic profiles can be produced
and overlaid on exploration fields, to determine both the geologic properties
and the risk of equipment damage from earthquakes once a production site is
active. The exploration fields can be given in one map layer, shown on the left
in Figure 1.5, that overlays another layer on the right which shows proposed
locations of well sites and pipelines. The full GIS incorporates both terrestrial
(topography, well sites, pipelines) and subsurface features (earthquake fault
zones, oil deposits).
Another upstream application displays the leasing of exploration land areas
held by GIO and its global competitors. It is crucial to know precisely the
boundaries of what land is leased compared to competitors’ leases and new
blocks that might be bid on.
Once drilling starts, 3-D map images can show how the exploration field is
laid out, both above and below the surface. These 3-D map images can be used
to plan, monitor, and evaluate the construction, regulation, and environmental
impacts of pipelines. Pipelines constitute GIO’s second largest spatial use
after geologic modeling.
Figure 1.5 Map Overlay of Exploration Field Topography and Locations of Pro-
posed Pipelines and Well Sites
Figure 1.6 GIS in GIO’s Retail Market Planning in Africa
For transportation and storage, multi-layered maps show routes, depots,

and associated supply districts. For marketing, service station locations are
analyzed relative to retail markets, price zones, urban and population areas
from the census, and trade area maps.
An example of retail market planning is given in Figure 1.6 for a retail area
in Africa. Map layers show major roads and service stations in a suburban area,
suburban boundaries, service stations in a denser urban area, and provincial
boundaries. Based on these, GIO managers can make improved decisions on
locations of retail stations that take maximum advantage of traffic flows, and
population proximities, and distribution of existing stations.
GIO has critical supply-chain models and information systems that support
the planning and movement of items worldwide. The company has begun to
add spatial analysis and mapping to its supply chain system. Figure 1.7
indicates routing pathways between plant and container locations for Europe,
Africa, South Asia, and Australia. Questions can be answered such as what are
the distances to ship items, how expensive are routes, and how can shipments
be best timed and coordinated over space-time to create the most efficiency.
In the environmental realm, GIS is applied upstream and midstream to
model layers of physical topography, hazards, and social impacts. Strategic
planning uses include spatial analysis of global new ventures and merger
proximities of production facilities and office locations. These applications
spatially compare data on GIO’s facilities, workforce, and assets to those of
new-venture or merger candidates. Is there a fit geographically between GIO
and other entities? Can the proposed consolidations be beneficial? What are
the savings from proximities?
Figure 1.7 GIO’s Supply Chain Routing of Containers from Plant to Customer in
Africa, South Asia, and Australia
Application of GIS to company strategy at GIO can be analyzed through

matrices, such as the simple one illustrated in Figure 1.8 categorizing certain
parts of the holding company and its many subsidiaries where GIS/Spatial is
applied. In this matrix, the life-cycle stages are the columns, while the rows
represent functional business sectors. The number of stars reflects the amount
of GIS application. It’s clear that GIS finds its way into most cells. At GIO,
GIS has moved far from a “silo” and diffused widely into most of the business
sectors, life-cycle stages, and subsidiary companies.
Given GIS’s prevalence, the company has a surprisingly small GIS Cen-
tral Group of a dozen experts, located in one of the subsidiary technology
companies. That group interacts with twenty-five other GIS employees who
are scattered in many of the dozens of GIO operating companies. The GIS
Central Group does management, planning, and provides the expert spatial
consulting for all the companies, as well as coordinating with the spatial
employees worldwide.
The company is pushing the technology limits, in areas such as sub-surface
geologic mapping, web-based applications, 3-D maps, data warehousing,
and advanced spatial analytic tools for business planning. Overall, GIS is
considered by GIO top management to be very strategic in taking on GIO’s
competition at many levels.
GIS through the Business Lifecycle

by Business Sectors
Business
Lifecycle
Opportunity Operations Asset

Development Disposition
Exploration
and
Production
Business Transportation
Sectors and Storage
Refining
Marketing
Support
Figure 1.8 Matrix of GIS Applications at GIO by Business Sector and Life-Cycle
Stage
THE SPATIAL RE-EVOLUTION
In the past few years, GIS and spatial developments in business have come
at a rapid pace, leading to “re-evolution” of earlier business paradigms for
GIS. This section looks back at key milestones and then considers several
waves of new developments sweeping across GIS and spatial technologies
that include RFID, sensors, Lidar, spatial web services, open source software,
and event-driven architecture. This section sets the basis for some technology
themes of the book.
Historical Antecedents
The history of GIS and spatial technologies in business goes back only about
fifty years and only in a major way since the 1990s. However, antecedents
go back to non-computerized map overlays in the early twentieth century. A
spatial overlay is the process of exactly superimposing of several map layers
such as terrain, hydrology, geology, and agriculture. The layers can be viewed
together, and the relationships of their features analyzed (Wade and Sommer,
2006). Using exact scales to match up layers, Jacquiline Tyrwhitt did this
in her Town and County Planning textbook in 1950 (Clarke, 2003). Waldo
Tobler’s article in 1959 conceptually foresaw using a computer to accomplish
cartography from inputting data to producing maps (Tobler, 1959). In the
THE SPATIAL RE-EVOLUTION 15
Figure 1.9 ESRI Inc. Founder and President, Jack Dangermond. Source: Copyright
© ESRI. All rights reserved. Used by permission
1960s early computer programs such as SYMAP and CALFORM could

overlay layers and produce maps using mainframe computers and on bulky,
crude printers. The Canadian government was an early adopter in 1964 of
primitive GIS and continued in subsequent decades to be a lead user. An
academic milestone was the book Design with Nature by Ian McHarg, a
professor at University of Pennsylvania. McHarg included plastic overlay
sheets that demonstrated how overlays could shed new light on environmental
design (McHarg, 1969).
In the mid to late 1960s, substantial progress was achieved on computer
algorithms for GIS at the Harvard Laboratory for Computer Graphics and
Spatial Analysis, including the use of arcs and nodes to build up boundary
files (Clarke, 2003). In the late 1960s, Environmental Systems Research In-
stitute (ESRI) was formed as a company and headed since its founding by
Jack Dangermond (see Figure 1.9). Its mini-computer product ArcInfo, based
on Harvard concepts, became the leading commercial GIS software package.
Although ESRI’s early markets were predominantly in government, this also
marked the beginning of GIS and spatial applications for businesses. Later
other software companies entered the markets, including MapInfo Corpora-
tion and Intergraph Corporation. During the Cold War, there were parallel
and often secret developments in mapping and satellite technology by U.S.
military planners who were reacting to the threat of nuclear war (Cloud,
2002; Charles, 2005). This stream of military R&D eventually led to the
first development and deployment of satellite systems for GPS in the 1980s.
The military history is less well known and was largely detached from the
academic developments of those decades.
For the 1970s through the 1990s, GIS platforms progressed from bulky
mainframes to mini-computers, pcs, and handheld devices. User interfaces
changed from command-driven to Graphical User Interfaces (GUIs) in the
1980s. These transformations were driven by advances in the computing
TABLE 1.2 Moore’s Law. Transistor Capacity of Intel Processor Chips, 1971–2006
Year of Introduction Chip No. of Transistors per chip MIPS∗
1971 4004 2,250 0.06

1972 8008 2,500
1974 8080 5,000 0.64
1978 8086 29,000 0.75
1982 286 120,000 2.66
1985 386 275,000 5.00
1989 486 1,180,000 20.00
1993 Pentium 3,100,000 66.00
1997 Pentium II 7,500,000 1,000.00
1999 Pentium III 24,000,000 1,354.00
2000 Pentrium IV 55,000,000 9,700.00
2002 Itanium 250,000,000
2003 Itanium 2 600,000,000
2006 Dual-Core Itanium 2 1,100,000,000 27,000.00
∗ millionsof instructions per second
Source: Intel, 2007.
power of the underlying component technology. There was a 450,000-fold

increase in chip capacity and in processing power from the 1971 through
2006 (Intel, 2007) (see Table 1.2).
The expansion in power allowed GIS to run on much smaller devices
or to run thousands of times more powerfully on the same device. Other
technological advances impacting GIS were vastly improved monitor screens,
higher-resolution printers, expanded networking lines and bandwidths, and
more advanced telecommunications design and protocols.
GIS software, most of which was originally written in FORTAN, pro-
gressed in the 1990s to object-oriented (OO) languages such as C++ and Vi-
sual Basic. Object-oriented design emphasizes interchangeable components,
interactive GUI interfaces, and greater ease of use. OO design also allowed
GIS software to be more portable across computing platforms. As a result
today’s GIS software can be implemented as a family of products with the
similar user interfaces extending from handheld devices to large-scale servers.
Networking changes in the late 1980s and 1990s transformed the distribu-
tion and reach of GIS. For large-scale applications, client-server architecture
was introduced on conventional lines (non-internet). Most of the processing
takes place on the server(s), which connect and share processing with the
client’s local system. Data can be stored on the central server, on special-
ized database servers, and/or on the client’s system. Also, during this period,
versions of all the leading software packages such as ESRI’s ArcInfo and
Intergraph’s Geomedia were introduced for client-server platforms. Today,
powerful client-server versions support advanced departmental applications
often for specialized purposes. Client-server architecture was not designed

for the internet whereas enterprise web services is. As the latter has be-
come more prevalent, the client-server environment can be referred to as
“traditional” (Sonnen and Morris, 2005), even though it is still in widespread
use. The distinction of “traditional” versus enterprise/web-based will carry
through the book as a theme.
Coupling Technologies
Coupling technologies such as GPS, RFID, and portable wireless devices
became prevalent in the 1990s. A coupling technology is one associated with
GIS that makes the combination more productive and efficient (see Table 1.3).
For example, GIS by itself can produce maps based on existing datasets, but
does not have the capacity to gather data in real-time. However GIS coupled
with GPS can gather geo-referenced data, input it, combine it with existing
datasets and produce real-time outputs. As another example, GIS in a field
location formerly functioned on a desktop computer with dial-up networking.
It is replaced by GIS coupled with a light, flexible, and physically trans-
portable wireless device that is networked at higher bandwidth. The handheld
computer in Figure 1.10 is supported by the ArcPad software from ESRI. It
provides features for map viewing, data query, distance measurement, GPS
navigation, data editing, limited spatial analysis functionality, and wireless
communications. Data can be transferred back and forth between the ArcPad
mobile device and a networked workstation with GIS by wireless, Bluetooth
connections or a physical cable.
A list of some of the coupling technologies is given in Table 1.3.
Global Positioning Systems (GPS) technology, developed in the U.S. mil-
itary beginning in 1973 and utilized for the first time in combat during Op-
eration Desert Storm in 1991, was not made available for commercial use
until shortly after the end of the Cold War in 1993 (Pace, 1995). However,
the military at first broadcast a civilian signal that was only accurate to within
100 feet (NAS, 1997). In 1996, the Clinton Administration approved a higher
level of GPS accuracy within several feet to be made available to anyone
(NAS, 1997). GPS is based on a system of twenty-four Navstar satellites that
were constructed by the Rockwell Division of Boeing (originally Rockwell
International) and completely deployed in 1993. Each satellite completes a
full low earth orbit every 12 hours. They are arranged in an orbiting forma-
tion so every point of the earth is in radio communications with at least four
satellites at all times (NAS, 1997). GPS devices receive signals from four of
the satellites that enables the device to determine its position anywhere in the
world with a very high accuracy, commonly under one meter. Military GPS
have even higher accuracies as low as a centimeter.
Radio Frequency Identification (RFID) technology started earlier than
GPS. Invented by Léon Theremin in the Soviet Union in 1945, it was pre-
sented conceptually in a paper by Harry Stockman in 1948 (Landt, 2001), and
Figure 1.10 Handheld Computer Running ArcPad Software. Source: ESRI Inc.
further developed and tested in the 1970s and 1980s. RFID allows automatic
identification of physical items which have tags (active or passive) that con-
nect with RFID readers. Passive tags can be read within about twenty feet,
while active tags can be read within several hundreds of feet and have larger
memories. Its first commercial applications were in the late 1980s. It became
widely deployed the 1990s after standards were developed and agreed upon
(Landt, 2001). In the past few years, lead organizations such as the U.S. De-
partment of Defense and Wal-Mart Inc. mandated its use for suppliers, and
their “clout” has speeded up business adoptions. An RFID reader records the
data on an item at a particular time and place, resulting in geo-referencing to
the reader. It is not continually referenced, as with GPS devices, but occurs
at discontinuous points when a tag goes by a reader. The advantage of RFID
is that huge numbers of items can quickly be geo-referenced, inventoried,
spatially analyzed, and mapped.
An RFID device has a processor, memory, and a radio antenna. Passive
RFID devices do not transmit but can be read by radio signals in range of up
TABLE 1.3 Technologies Often Coupled with GIS
Technology Importance for GIS in Business
GPS Global Positioning Systems (GPS) is a technology that can

determine exact point locations anywhere on the earth by
communicating signals with four of the twenty four GPS
satellites in an orbiting system. GPS combined with GIS allows
real-time locational information to be mapped for business
purposes.
RFID Radio Frequency Identification (RFID) allows portable products of
any type to be spatially registered and to carry data that can be
accessed and updated remotely by RFID readers. Its use in
business is that supply chains and inventory items can be
RFID-tagged and their positional locations tracked and stored
for GIS access (Richardson, 2003).
Sensors Devices implanted into buildings, transport vehicles, inventory
items, the landscape, and even people provide physical
measurements of the environment, such as temperature,
lighting, radiation, noise, heat, etc.
Mobile wireless Allows field deployment of GIS technologies in mobile
communications commerce. Useful in supporting the real-time field operations of
businesses (Mennecke and Strader, 2003). Combines GIS, GPS,
and wireless technologies.
Handheld Devices Portable handheld spatial devices, such as Personal Device
Assistants (PDAs), cell phones, handheld computers, and other
mobile devices. For spatial purposes, it contains a GPS device
and scaled-down versions of standard GIS software. Gives
businesses field flexibility in inputting, outputing, modifying,
analyzing, and utilizing data. Important in business sectors such
as retail and utilities that have substantial field forces.
LIDAR LIDAR (Light Detetion and Ranging) can estimate precisely the
distance to a surface or object by bouncing laser pulses. The
time delay from transmission to receiving the light signal
determines the distance. Utilized in aircraft and ground stations
for high-precision distance measurements.
to about 20 feet from an RFID reader. Active RFID devices actively transmit
radio signals so the readers can be located at more distance—up to 300 feet.
The costs of the tags and readers are falling rapidly. Readers have become
quite affordable, many models at less than $1,000, and are widely distributed.
As an example of the importance of RFID, Wal-Mart in 2006 required its
largest 300 suppliers to have RFID tags on its shipping crates and pallets.
Each RFID tag stores an Electronic Product Code (EPC), which provides
more information than the traditional bar code. In particular it provides codes
not only for product description but also a code for the production lot at the
time of manufacturing. This means that an item being shipped or stored can
be associated with its detailed production, transport, and storage history. The
transport and storage is the accumulated history of movements through read-
ers, which is transmitted and stored in a database. Wal-Mart will track crates
and pallets from the point of entering the Wal-Mart supply chain through
distribution centers to individual stores (Williams, 2004). Whenever the item
passes an RFID reader, not only the EPC but also the coordinate location of
the reader is recorded. This supports mapping and spatial analysis applica-
tions in order to better plan and optimize the spatial movement and storage of
inventory and improve transportation and optimal locating of inventory items.
Portable wireless devices were introduced for widespread commercial use
in the 1990s and continue to expand in functionality and market penetration.
They include handheld computers, mobile phones, PDAs (personal device
assistants), and combined devices such as Blackberries. They support users
in the field for data input and display, for instance utility field maintenance
staff who need to view maps to locate and repair pipelines, control boxes, and
underground facilities.
Wireless LANs are used to provide networking in a flexible local envi-
ronment without cabling. It allows personnel to capture or update spatial
information into devices often portable ones, unfettered by cabling restric-
tions.
LIDAR creates vivid earth imagery that may be used for weather forecasting
and analysis, regardless of whether there is spatial analysis. Its displays are
useful for viewing earth events for business and government analysis and
decisions.
Networked sensors collected a variety of environmental information, usu-
ally at a fixed location. They can measure the magnitude and volume of of
environmental or human events at the sensor’s location. For instance, climatic
data from a network of weather sensors in a region can be input in real-time
into a GIS to produce continuous weather maps. Sensor-based data can be
used for supporting decisions in business and government (Meeks and Das-
gupta, 2005). In short, many of the Geo-location devices share spatial and
non-spatial uses.
Broadband as the Foundation for Web Services

Another shift was the advent of web services and service-oriented archi-
tectures over the past decade (see Table 1.4). The shift is supported by the
expanding internet capacities shown in Table 1.1 and by the widening band-
widths of internet transmission (Austin and Bradley, 2005). Larger bandwidth
implies that greater volumes of information can be transferred per unit of time
(Dodd, 2005). An important bandwidth division point occurs between nar-
rowband and broadband, which generally means bandwidths higher than that
of T1 transmission lines (Dodd, 2005). The dividing point is at 1.54 million
bits per second (mbps).
TABLE 1.4 Broadband Penetration Rates by Country
Households Using
Broadband Broadband as Percent Percent of Population
Subscribers Per 100 of Total Households with High-Speed
Rank Country Persons (1/2004) (12/2003) Internet Access (2002)
1 South Korea 24.0 70.5 21.3

2 Hong Kong 17.0 50.3 14.6
3 Canada 15.0 36.2 11.5
4 Iceland 14.5 NA 8.6
5 Taiwan 12.5 43.2 9.4
6 Denmark 12.0 25.7 8.6
7 Belgium 11.5 24.7 8.4
8 Japan 11.0 28.0 NA
9 Netherlands 11.0 NA 6.5
10 Switzerland 11.0 23.1 NA
13 United States 10.0 22.5 6.5
Source: Reprinted by permission of Harvard Business School Press. From The Broadband Explosion by
Robert D. Austin and Stephen P. Bradley (eds.), Boston, MA 2005.
Today’s broadband has capacity much higher than that, for instance cable
modem services support 6 mbps. T3 lines have capacity of 44.7 mbps, more
than the average individual user can take advantage of (Dodd, 2005). Broad-
band is rapidly becoming more prevalent, including several nations that had
more than 50 percent of households connected to broadband in 2003. Many
advanced nations had over 20 percent (see Table 1.4).
The growth rates of conversion to broadband are high, so it can be expected
within several years to be the dominant transmission mode in developed
nations. This trend favors spatial web services, which require rapid and high
volumes of information transfer, i.e. in the broadband range.
Spatial Web Services

Spatial web services are those that can be delivered to the user over the
web. They are delivered over the Internet Bus, which refers to the high-speed
internet or intranet available throughout the enterprise. As seen in Figure 1.11,
the organization can put its spatial software on an enterprise server, which
creates maps, performs spatial analysis, and sends outputs across the Internet
Bus. Also, database servers with data stores are connected to the Internet Bus.
Spatial web services managed by outside service vendors are linked to the
bus. The maps and other spatial output are transmitted over the Internet Bus to
reach users who have access to thin client systems, i.e. ones without much of
their own capacity, or thick client systems, ones that have their own software
packages and databases. At the lower left of the diagram is a connection to
business enterprise systems such as ERP, CRM, and Supply Chain.
Thick Client
Web Other Client Thin
Service Applications Applications
Applications
Internet Bus (Enterprise-Wide)
Business Spatial Web

Spatial
Enterprise Database Services
Enterprise
Systems Server from from
Server from
(ERP, the Firm Services
the Firm
CRM) Documents Vendor
Data Stores
Figure 1.11 Spatial Web Services Provided by Enterprise-Wide Internet Bus
This framework utilizes the protocol of Service Oriented Architecture

(SOA). SOA differs from traditional client-server protocols in supporting
a set of loosely coupled and interoperable application services that are inde-
pendent of the hardware platforms and programming languages. The most
common web services version of this is based on Simple Object Access Pro-
tocol (SOAP). It supports SOA as a web service based on the Extensible
Markup Language (XML) web protocol. SOAP allows application programs
to coordinate with each other across the Internet Bus, even though they uti-
lize different operating systems, programming languages, and technologies.
Thus it supports the integration of business applications, allowing sharing of
resources and lower cost.
Open Source Software

Another important trend is open source software, which has grown since
the late 1990s. Open source refers to software development by a community
of developers who contribute to publicly available software code. It may or
may not have copyright or patent protections, but if protected, the licens-
ing makes the software freely available. Probably the most famous open
source software is the Linux operating system, which has entered the main-
stream pc operating system market with a small but growing share. Open
source development is becoming more important for spatial and GIS soft-
ware, although no Linux-like software has yet become a leader. An example
is MapServer software, developed at University of Minnesota, which allows
development and rendering of maps for the web (MapServer, 2007). It’s not a
full GIS software but has many basic spatial features and tens of thousands of
users.
The open environment was also stimulated by the product and legal ap-
proach taken by the internet companies Google, Yahoo, and Microsoft when
in 2005 they released their free spatial software, Google Earth, Yahoo! Maps,
and Microsoft Virtual Earth. The firms opened up their code for outside de-
velopment, by releasing the APIs. An Application Program Interface (API)
is a group of protocols, tools, and code modules that allow developers to
build applications consistent with the original software, although the firm
may retain some limited controls. The outside programmer can use the API
elements as a set of building blocks to develop applications that enhance
or expand the original software. An example for GIS is a property devel-
opment company in Portland, Oregon, that utilizes the Google Earth API,
combined with its own property database to display satellite imagery of
its properties along with their value, tax status, square footage, and year
built.
Event-Driven Architecture
Along with SOA and open source, another web trend for GIS/Spatial is event-
driven architecture. Event-driven architecture (EDA) refers to developing
applications in which software events in certain combinations or sequences
trigger sending of messages in an asynchronous manner to software modules
at remote sites that are unaware the messages are coming (Silwa, 2003).
Asynchronous refers to transmission of messages at different times, while
synchronous is at the same time. It is “de-coupled,” in the sense that the event,
sender, and recipient are not previously connected. Middleware serves as the
intermediary to which event-messages are sent and allocates the messages to
software programs that have subscribed to be notified (Silwa, 2003). EDA
is a decoupled, asynchronous architecture that complements SOA’s loosely
coupled synchronous architecture.
In GIS and spatial business applications, an example of EDA is for a
transportation firm that sends out notifications when a vehicle is errant from
its course. It can set up a combination of event criteria that define an errant
vehicle, e.g. it is more than 5 miles outside its travel buffer zone and at a speed
more than 10 miles per hour above the speed limit for more than a half hour. If
the vehicle goes errant, the notification is automatically sent to a subscription
list of office managers, customers, and security personnel. A current weakness

of EDA is lack of widely accepted standards, which means that it runs best
where all the programs involved are linked to the same middleware (Silwa,
2003).
In summary, GIS and spatial technologies have advanced tremendously
over the past fifty years. Not only have capacities of the underlying component
technology increased vastly and the internet revolutionized approaches to
transmitting information, but new coupling technologies such as GPS, RFID,
LIDAR, and Sensors have extended the combined functionality. Many of the
key developments for business took place since the web’s advent in 1989,
so widespread adoption of spatial technologies in business is fairly recent.
Even if a technology diffuses rapidly, companies and their people are often
slower to adopt it due to prior investments, measured adoption approaches,
and resistance to organizational change.
The overall changes in the IT context of the evolution of GIS and spatial
technologies are summarized in Figure 1.12.
GIS parallels this IT timeline. It commenced in the Batch Era in the late
1960s and progressed through the Online Era of the 1980s and Distributed
Era of the 1990s and into the first decade of this century (IDC, 2005). The
year 2005 was a key turning point in GIS history, marking the start of the
map services era, with free dynamic online processes for the general public,
as typified by Google Earth and Microsoft Virtual Earth.
In summary, spatial technology trends have transitioned from mainframe
to client-server to web services and event-driven architectures; from larger
to smaller computing devices; from batch processing to interactive, real-time
processing; from command-driven user interfaces to user-friendly GUI ones;
from separated departmental data to data shared enterprise-wide or multi-
enterprise-wide; from lower to higher quality standards of spatial information;
and from traditional, stand-alone GIS software to GIS coupled with a group
of associated technologies.
THE SPATIAL INDUSTRY
This section considers the spatial industry, which.comprises GIS vendor firms,
distributors, data providers, spatial service providers, firms for the associated
technologies (GPS, RFID, LIDAR, sensors), consultants, and integrators.
Overall, the industry revenues are estimated in the range of $20 billion,
with a further estimated $40 billion in government spatial data collection
(based on Sonnen et al., 2005; Daratech, 2005; and Longley et al., 2005).
The main components of the industry may be divided into (1) GIS software
and add-ons, (2) geolocation technologies, (3) web-based enterprise-wide
applications and infrastructure, and (4) spatially-enabled enterprise database
products.
c01
2005
Paleo-Computing Scope of current IT Legacy Dynamic process support
JWPR043-Geo-Business
Online Era Services Era

October 18, 2007
Batch Era Distributed Era

13:20
25
1970
1980
1990
2000
Client / server
Web services
Online inquiry
Char Count=
Tape-to-Tape processing
Even driven architecture
Service oriented architecture
Source: IDC, 2005

Ubiquitous location determination
Figure 1.12 Evolution of the IT Context for GIS, 1965–2005. Source: Sonnen et al., 2005
TABLE 1.5 GIS/Spatial Software Worldwide Revenues from Leading Vendors
GIS/Spatial
Revenues, 2004, in Percent of Market, Annual Growth Rate
Vendor Firm millions of dollars 2004 2002–2004
ESRI 445.0 22.6 2.0

Intergraph 376.9 19.2 −0.3
Autodesk Inc. 261.3 13.3 24.1
MapInfo 97.7 5.0 15.4
AOL - Mapquest 67.0 3.4 6.1
Microsoft 56.0 2.9 16.7
GE Energy 18.8 1.0 −53.1
Group 1 Software Inc. 11.7 0.6 9.4
ObjectFX 10.0 0.5 11.1
PCI Geomatics 8.1 0.4 53.0
Other 612.2 31.2 5.6
TOTAL 1,964.7 100.0 5.6
Source: Sonnen et al., 2005
(1) GIS Software and Add-Ons

GIS revenues from the ten leading core software vendors in 2004, shown
in Table 1.5, reflects that four companies, ESRI, Intergraph, Autodesk Inc.,
and MapInfo control 60 percent of the market.
1. ESRI, the long-time and largest general-purpose software and services

vendor, introduced ArcGIS 9 in 2005 as its new lead product family.The
family offers GIS software ranging from enterprise applications at the
high end to software for portable devices, and including map server op-
tions. The products are built in object-oriented components that allow
them to function in a variety of environments. It licenses web server
software software to firms as ArcGIS Server. For enterprise-wide ar-
chitecture, ArcGIS is provided as component libraries residing on the
ArcGIS Server. Outsider developers can gain licensed access to the
ArcObjects components through ArcGIS Desktop, Engine, and Server.
ESRI also has divisions that provide selective and large-scale appli-
cations consulting as well as management of a client’s web services
under ArcWeb Services and other offerings. ESRI has been a technol-
ogy leader in GIS software design. ArcPad software is offered to mobile
users, and Business Map and Business Analyst Online to medium- and
small-sized enterprises.
2. Intergraph Corporation provides software to middle- and large-sized
government organizations and businesses. It also sells hardware to
support the software. It focuses on vertical markets including homeland
THE SPATIAL INDUSTRY 27
security, government, power and utilities, manufacturing, and trans-

portation. Its leading product family, Geomedia, is offered in web,
client-server, and desktop versions.
3. Autodesk Inc. especially serves the utility, manufacturing, building, and
infrastructure sectors, and partners with Oracle for enterprise-based
applications. Its software has tools for engineering, design, planning,
and managing large land-based assets. Autodesk heavily serves the
utilities industry, where engineering exactness and asset tracking is
essential. It extended its products into location-based services to support
small and middle-sized businesses in their mobile workforce and field
operations. Its major client base comprises government organizations
and large infrastructure-related firms. Autodesk grew rapidly from 2002
to 2004, the result of its emphasizing the areas of homeland security and
government infrastructure that benefited from the post 9/11 build-up.
4. MapInfo markets software to users ranging from individuals to medium-
sized organizations. Its lead product is the full-featured GIS software
package, MapInfo Professional. Other products are in enterprise appli-
cations, geodemographics, and street mapping data. Mostly it sells to
medium-sized businesses and emphasizes marketing, banking, insur-
ance, utility, and government applications. MapInfo, which since 2007
is a division of Pitney Bowes, has not yet introduced data management
engines, map servers, or mobile device products. It is sometimes known
as PB MapInfo.
The GIS software industry also has hundreds of smaller, more-specialized

firms that often work in partnership with leading vendors. An example is PCI
Geomatics (number ten in the Table 1.5), a Canadian firm that offers services
in remote sensing, digital photogrammetry, terrain analysis, radar analysis,
and data visualization software development, as well as mapping products to
support these areas.
All told the GIS software industry in 2004 totaled about $2 billion in rev-
enues (IDC, 2005). IDC estimates that the segment of industry that provided
associated services in 2004 was $8 billion (IDC, 2005). These firms provide
design, development, consulting, facilities management, and outsourcing to
implement, manage, and enhance the core software.
The industry also includes a segment of systems integrators, software
distributors, and software retailers. There are many partnerships and inter-
relationships among these enterprises (Daratech, 2004), as seen in Figure 1.13.
Daratech estimated the total GIS industry and services in 2004 at $8.5
billion, with a somewhat lower services component of $6 billion. It includes
specialized spatial hardware vendors with revenues in 2004 of $650 million
(Daratech, 2004). An example of a specialized hardware vendor is Trimble
Inc. which manufactures and markets a variety of small- to large-sized GPS
devices.
RESELLERS
($1.8 Billion) FRONT-END
BUSINESS
BACK-END ($8.5 Billion)
BUSINESS FRONT-END CORE
($738 Million) VENDORS
Systems
Integrators ($1.8 Billion)
Software
Developers HARDWARE END
(Non Core) USERS
Hardware $650 Million
Manufacturers
Service
SERVICES
Providers
(Non Core)
$5.5 Million
Revenues are
Estimated for 2003.
Source: Daratech Inc., 2003.
Figure 1.13 Model of the GIS Marketplace, 2003
Systems integrators support the planning and integration of the groups of

spatial and other technologies. They establish processing flows, data flows,
and software and networking coordination, when varied hardware and soft-
ware products are applied in combination. This is often in conjunction with
data providers and networking platforms. Systems integrators are especially
valuable for large-scale projects, which must realize complex configurations
of hardware, software, and applications. Here, the systems integrator will be
present throughout the project to advise management. Among the steps in a
large project that the integrator consults on are the feasibility study, conceptual
design of hardware and software, requests for proposals (RFPs), specifica-
tions, quality assurance, developing, and building interoperability between
different vendors, implementation, and training (Daratech, 2005). In sum, the
integrator acts in a high-level technical and management consulting role.
If the project management is fully delegated to an outside party, then
outsourcing occurs. Although not yet prevalent for GIS, outsourcing is on the
rise and is covered in Chapter 10. The integrated arrangement of GIS and
associated components, human resources, and processes needs to be stressed,
and will be seen throughout the book and its cases.
(2) Geolocation Technologies
These technologies, GPS, RFID, wireless LANs, cellular networks,
Lidar, Sonar, and networked sensors (Sonnen et al., 2005) have already been
introduced. They advanced greatly in the present decade and new ones are
likely to be introduced (Sonnen et al., 2005). Many smaller, specialized firms
THE SPATIAL INDUSTRY 29
in the spatial industry and some larger ones provide and support these devices.
Since these technologies are heavily utilized in military and intelligence, this
side of the spatial industry is allied with the defense industry.
(3) Web-Based Enterprise-Wide Applications and Infrastructure
Many businesses have already developed IT infrastructures that support
enterprise applications such as Enterprise Resource Management (ERP), Cus-
tomer Relationship Management (CRM), Data Warehousing, Supply Chain
Management (SCM), and Electronic Commerce (Gray, 2005). The develop-
ment of these systems was stimulated by the Year 2000 necessity to re-design
large legacy systems to avoid the feared failure of built-in clock algorithms
in many of them. Another stimulus to such systems was the more robust web
and the trend of the underlying networking towards higher bandwidths.
Vendors of many of these large-scale business systems are adding spatial
modules or interfaces to GIS software. An example is the ERP leading ven-
dor SAP, which has introduced spatial connectors. In 2006, there were five
technical interfaces that connected ESRI ArcGIS and ArcIMS software with
SAP (ESRI, 2006):
(1) SAP remote function calls (RFCs) between SAP and ArcGIS
(2) third-party connectors
(3) the SAP GIS Business Connector
(4) enterprise application integration from both SAP and third parties
(5) direct connections in prototype stage between SAP and ESRI software
Each of the connectors has pluses and minuses. For instance, the Belgian util-
ity firm Pidpa uses ArcIMS to create a website called GeoLink that connects
ERP and GIS data. The application is for utility monitoring and mainte-
nance applications known as SCADA. Pidpa is examined as a case study in
Chapter 4.
(4) Spatially-Enabled Enterprise Database Products
Enterprise database software has spatially-enabled versions that create a
platform strong on database functionality, yet providing extensive spatial
functionality as well. This approach to GIS emphasizes seamless incorpora-
tion of spatial functions into the backbone databases of the enterprise (Lopez,
2005). It is particularly suitable for large-scale applications involving ex-
tensive data and huge user bases across large organizations (i.e. millions of
concurrent users). Since that profile is already being well served by enterprise
databases for many corporate functions, using the same database with a spa-
tial extension is appealing. These large scale databases tend to be supported
by the IT department, not a separate GIS department. It is spatial with an
IT flavor to it, emphasizing huge processing and efficiency (Francica, 2005;
Lopez, 2005).
Two major database providers, Oracle and IBM, have spatially-enabled
database products. In the general database software marketplace in 2004,
Oracle had 34 percent of the market, followed by IBM at 29 percent (Stan-

dard and Poors, 2006). Since the database market leader is Oracle, it is not
surprising that the firm has led in introducing versions of the database software
with spatial functionality.
The recent version of the Oracle database enhanced with spatial func-
tions is Oracle Spatial 10g (Francica, 2005; Lopez, 2005; Oracle, 2005). This
database has moderate GIS functionality. Although the GIS functionality is
somewhat less than for mainstream GIS software, Oracle Spatial 10g of-
fers potentially greater integration with enterprise software and applications,
especially for firms that have already adopted Oracle as a standard. How-
ever, for certain areas such as business intelligence and ERP, Oracle Spatial
10g’s functionality may exceed standard GIS software (Francica, 2005). The
mainstream IT connectivity in the enterprise arena makes Oracle Spatial 10g
competitive with traditional GIS software offerings for high-volume, huge-
user-base applications. Companies can even have both, Oracle Spatial 10g
underneath for generalist applications and one or more traditional GIS soft-
ware packages running on top for particular specialties (Lopez, 2005). There
are also service and consulting firms that provide software add-ons, services,
and integration for the web-based Enterprise Applications. The architecture
of spatial database applications are compared further in Chapter 8.
Overall, the emerging spatial market includes desktop uses of GIS, tra-
ditional client-server GIS software, and web-based enterprise applications
(Sonnen et al., 2005). As seen in Figure 1.14, the established GIS market has
a high spatial focus and medium to high integration complexity (Sonnen et
al., 2005). “Integration complexity” refers to how complex it is to integrate
GIS with other systems and applications. Established GIS is an older seg-
ment that is informed by a lot of geographical knowledge and a strong and
High
Enterprise Location Services and Software
Integration Established
Complexity GIS
Low Personal Use
High Geospatial Low

Focus
Figure 1.14 Model of Geospatial Markets in 2004. Source: Sonnen and Morris,
2004
STRATEGY, COMPETITIVE ADVANTAGE, AND GIS 31
dedicated workforce of GIS professionals, but less expansive architecture. A

second part of the GIS market consists of spatial software for personal use
that may be implemented on desktops, laptops, and PDA devices. Its spatial
focus varies from high to low. Users range from traditional GIS experts to
emerging business users. Its integration complexity is low, but it offers user
convenience and flexibility. The emerging market for web-based enterprise
applications, seen in the upper right, serves users with mostly middle to low
spatial focus, and the applications are medium to high in integration com-
plexity. The most rapid growth in businesses is for the latter group of users,
as a number of the case studies will demonstrate.
STRATEGY, COMPETITIVE ADVANTAGE, AND GIS
Companies are successful or unsuccessful in IT depending on how they

set up business models, deploy strategies to take advantage of their core
competencies, and move ahead of the competition (Applegate et al., 2007).
Successful firms need to understand and plan for (1) strategic opportunities
and (2) strategic risks (Applegate et al., 2007). Opportunities exist at all times,
but are accentuated when firms and industries are undergoing rapid changes,
such as happened in the Dot.Com era of the late 1990s. Earlier sections of
the chapter have indicated that the early twenty-first century is a time of
rapid change and innovation for GIS and spatial technologies in business.
The current pace of change is seen by the advent in the past five years of
the enterprise web platform for GIS; the advent of widespread RFID uses
in firms and industries, often combined with spatial analysis; the appearance
of free consumer web packages such as Google Earth; and the increasing
sophistication of the large-scale spatial databases such as Oracle Spatial.
As in earlier times of change, successful firms will spend time to assess
risks and come up with strategies that move the company forward versus its
competition. Although the details of strategic and competitive GIS/Spatial
are covered in Chapter 12, this section briefly considers the strategic grid and
spatial strategic alignment models, and examines them in the context of GIS.
The latter is a spatial strategy model for the current period of rapid change
and innovation. It underscores why the web integration platform is important
in understanding spatial strategies. Sperry Van Ness, a commercial real estate
firm, exemplifies as a case how a competitive web strategy for GIS can be
successfully implemented.
Two models of strategy are considered.The McFarlan Strategic Grid, seen
in Figure 1.15, encompasses the two dimensions of (1) impact on operations,
and (2) how strategic it is for the business. The grid can be applied to GIS ap-
plications, which may or may not impact business operations and strategy. For
instance, Global Integrated Oil’s GIS depends on spatial analysis to model its
upstream exploration, especially the geological aspects of petroleum deposits
and exploration. For GIO’s geologic spatial analysis, it’s highly strategic and
it has a strong impact on operations, i.e. the firm’s global explorations could
High
Charles
Schwab
Nasdaq
Impact on Business Operations
Factory Strategic
Support Turnaround
Manufacturing
Firm
Consulting
Low Firm
Low High
Impact on Strategy
Figure 1.15 McFarlan’s Strategic Grid. Source: Reprinted by permission of

Harvard Business Review. This article was originally published under the title, “In-
formation Technology Changes the Way You Compete,” by F. Warren McFarlan in
May/June 1984. Copyright © 1984 by the Harvard Business School Publishing Corp.
not be done well today without it. It’s clear that GIO’s application falls into
the Strategic Grid’s upper-right cell, i.e. GIO can’t get along without it. This
model is returned to in Chapter 12.
The second model (Hagel and Brown, 2001) offers a way to organize
thinking about spatial strategy. It reasons that the web services architecture
increases strategic competitiveness. Firms that have tended to shed their for-
mer proprietary and closed approaches to the internet and instead developed
interaction with multiple other firms through the web and internet can real-
ize net benefits. Web services can be accessed globally, to combine the best
data and software applications available that are affordable and function well
together (Hagel and Brown, 2001).
The final chapter will demonstrate through testing with all twenty research
cases in the book that the presence of a web integration platform is strongly
associated with how strategic GIS is to the company. The implication is that
management of firms that can deliberately champion and effectively move for-
ward in the new spatial web environment will tend to do better competitively.
SPERRY VAN NESS: GAINING COMPETITIVE ADVANTAGE FROM GIS
Sperry Van Ness, a private firm founded in 1987, is a rapidly growing com-
mercial real estate brokerage firm with over 1,200 employees and commercial
SPERRY VAN NESS: GAINING COMPETITIVE ADVANTAGE FROM GIS 33
projects throughout the United States. It advises clients and brokers on over
a billion dollars annually in retail, industrial, office, multi-family, and land
sales transactions. The company has deployed web-based spatial applications
that support marketing and sales efforts by its national sales force of 800 peo-
ple, who are referred to at Sperry as “Advisors.” Sperry’s Advisors represent
the sellers and buyers of commercial properties that range in price from $3
million to $50+ million. Usually the owners are sellers but sometimes are
buyers. The top Advisors are hard working, highly paid at often well into six
figures, and critical to Sperry’s success as a company.
The GIS was developed over the past four years by a manager and team of
three technical specialists, a group also responsible for Sperry’s IT. The team
followed a non-traditional approach for GIS, adopting Microsoft software
including MapPoint and MS Virtual Earth on a web platform available both
on the company’s internet and intranet. GIS user training was emphasized
and offered online through the commercial Raindance Conferencing service,
which provides combined web and phone training conferences to the Advisors
and Advisor Support Center (ASC) staff that can be flexibly deployed across
time zones.
The goal of the GIS was to provide the Advisors and Brokers with maps,
photos of properties, property sales descriptions, and relevant demographic
information in a very user-friendly format. This allows them to showcase
online or in person the properties for sale and “tour” prospective investors on
the web around sets of prospective properties. For face-to-face meetings with
clients, Advisors can use the system to very quickly prepare fancy printed sales
material in person. Sperry Van Ness is unique among national brokerage firms
in implementing a broker marketing plan along with its investor marketing
plan. Its ethos is to encourage cooperation of parties engaged in commercial
real-estate transactions. Its three main spatial systems are the following:
1. General public viewing website. A publicly accessible part of the cor-

porate site that lists the firm’s properties for sale nationwide, it enables
the user to locate properties on a map, see a photo of the property, get
basic sales data and the building history, and examine supporting de-
mographic data. An example of the property description and mapping
web pages for a similar service is shown in Figure 5.4.
2. Online Publisher, a proprietary and fully featured software package that
serves the Owners, Advisors, and the Advisor Support Center (ASC)
staff, includes all features from the public site, plus report generation
in a highly polished look. Further, it allows the Advisor to “see” out
of the property at oblique views, i.e. at 30-degree angles. For any one
property, the user can select five viewing angles, a capability called
“immersive, birds-eye imagery.” The multi-view perspective allows the
broker, prospective buyer, and Advisor to imagine themselves inside
the property and get a sense of surrounding views, obstructions, and
surrounding neighborhoods. This unique feature, as well as the 1-hour

speed in putting together finished packages, sets Sperry apart from the
competition. In commercial real estate, speed of response is often of the
essence, especially for sought-after properties.
This case demonstrates a firm that is located in the upper-left strategic cell
of the McFarlan Strategic Grid, i.e. highly operational and highly strategic.
Since Sperry’s approach is totally based on an enterprise-web platform, it is
correlated with strategic competitiveness as postulated by Hagel and Brown
(2001). The firm’s top management is enthusiastic about the use of GIS, even
though they maintain a practical one- to three-year planning horizon on all
facets of the company including GIS.
CHAPTER SUMMARY
Geographic Information Systems emerged as an important factor in the busi-

ness world over the past ten years. The GIS trends are towards smaller devices,
coupling of GIS with associated technologies, web platforms, and enterprise-
wide applications. Some huge firms have adopted GIS across their enterprise,
as illustrated by Global Integrated Oil, which uses it throughout its value
chain and strategically in the upstream exploration and pipeline stages.
GIS and spatial technologies can be strategic and advance firms versus
their competitors. Some background on strategy is presented, including the
strategic IT grid and web services strategy. The strategic importance of GIS is
linked to use of enterprise-web platforms versus traditional client-server. The
case study of the medium-sized Sperry Van Ness commercial real estate firm
illustrates how well-conceived, user-friendly design in a web environment can
be very strategic in providing crucial maps and multimedia sales information
to top sales people much faster than the competition.
REFERENCES
Applegate, Lynda M., Robert D. Austin, and F. Warren McFarlan. 2007. Corporation
Information Strategy and Management, 7th Edition. Boston: McGrawHill-Irwin.
Ashley, Steven. 2003. “Next-Generation GPS.” Scientific American 289(3): 34.
Astroth, Joe. 2005. “LBS—from Killer App to Killer Enabler.” Directions Magazine,
September 4. Available at www.directionsmag.com.
Austin, Robert D., and Stephen P. Bradley. 2005. “The Broadband Explosion.” In
Austin, Robert D. and Stephen P. Bradley (eds.), The Broadband Explosion,
Boston, MA: Harvard Business School Press, pp. 3–20.
Barnes, Scottie, and Heather Gooch. 2004. “Tracking Orkin’s Bottom Line.” Geospa-
tial Solutions, May 1. Available at www.geospatialsolutions.com.
Bisio, Ron. 2005. “Mobile GIS.” GeoWorld, November.
REFERENCES 35
Boar, Bernard H. 2001. The Art of Strategic Planning for Information Technology,
2nd Edition. New York: John Wiley and Sons.
Boas, Nancy. 2004. “Mapping ChevronTexaco’s Memory.” CVX First Quarter,
pp. 16–21.
Bossler, J.D. 2002. “An Introduction to Geospatial Science and Technology.” In
Bossler, J.D. (ed.), Manual of Geospatial Science and Technology, London: Taylor
and Francis, pp. 3–7.
Brewin, Bob. 2005. “No Silver Bullets.” Federal Computer Week 19(13): 39.
May 2.
CCSF. 2007. “GIS Education Center: Providing GIS Solutions for Industry and
the Community at City College of San Francisco.” Office of Workforce and
Economic Development, City College of San Franciso. Available at ww.ccsf/
resources/geographic information systems, June 13, 2007.
Charles, Daniel. 2005. “Do Maps Have Morals?” Technology Review (June) 108(6):
77–79.
Choi, S.Y., D.O. Stahl, and A.B. Whinston. 1997. The Economics of Electronic
Commerce. Indianapolis, IN: Macmillan Technical Publishing (p. 626 has the
model of e-commerce market areas).
Clarke, Keith. 2003. Getting Started with Geographic Information Systems. Upper
Saddle River, New Jersey: Prentice Hall.
Cloud, John. 2002. “American Cartographic Transformation during the Cold War.”
Cartography and Geographic Information Science, July.
Crosswell, Peter L. 2004. “Know the Past to Understand the Present and Future.”
GeoWorld, December, pp. 32–35.
Dangermond, Jack. 2006. “GIS: the Framework for Information Management.”
Geospatial Solutions 16(1): 18–19.
Daratech. 2004. Geographic Information Systems Markets and Opportunities.
Cambridge, MA: Daratech.
Dodd, Annabel Z. 2005. The Essential Guide to Telecommunications, 4th Edition.
Upper Saddle River, NJ: Prentice Hall.
Francica, Joe. 2004. “MetaCarta Inc.—Geographical Text Searching.” Directions
Magazine, March 11. Available at www.directionsmag.com.
Francica, Joe. 2005. “LBS: Are Consumers Ready to Blow This Market Open?”
Directions Magazine, August 6. Available at www.directionsmag.com.
Francica, Joe. 2005. “Oracle Articulates Strategy to Special Interest Group.” Direc-
tions Magazine, March 30. Available at www.directionsmag.com.
Galliers, Robert D., and Dorothy Leidner (eds.). 2003. Strategic Information Man-
agement: Challenges and Strategies in Managing Information Systems. Oxford:
Butterworth-Heinemann.
Geoplace.com. 2005. “Wireless Carriers Extend E911 Deadline.” November 21.
Available at www.geoplace.com.
GeoWorld. 2005. “Industry Outlook 2005.” GeoWorld, December, pp. 24–31.
Gewin, Virginia. 2004. “Mapping Opportunities.” Nature 427:376–377.
Gray, Paul. 2006. Manager’s Guide to Making Decisions about Information Systems.
Hoboken, NJ: John Wiley and Sons.
Greenman, Catherine. 2000. “Turning a Map into a Layer Cake of Information.” New
York Times. January 20.
Hackbarth, Gary, and Brian Mennecke. 2005. “Strategic Positioning of Location
Applications for Geo-Business.” In Pick, James B. (ed.), Geographic Information
Systems in Business, Hershey, PA: Idea Group, pp. 198–210.
Hagel III, J., and Brown, J.S. 2001. “Your Next IT Strategy.” Harvard Business
Review, October, 105–113.
Intel. 2003. “Moore’s Law.” December. Retrieved from www.intel.com.
Sonnen, David, Henry D. Morris, and Dan Vesset. 2005. “Market Analysis: World-
wide Spatial Information Management 2005–2009 Forecast and 2004 Vendor
Shares.” Framingham, MA: International Data Corporation.
Landt, Jeremy. 2001. Shrouds of Time: The History of RFID. Pittsburgh, PA: Asso-
ciation for Automatic Identification and Data Capture Technologies.
Levy, Steven. 2004. “Making the Ultimate Map.” Newsweek, June 7, pp. 56–58.
Longley, Paul A., Michael F. Goodchild, David J. Maguire, and David W. Rhind.
2005. Geographic Information Systems and Science. Chichester, England: John
Wiley and Sons.
Lopez, Xavier. 2005. “Business and Technology Trends Driving the Future of Geospa-
tial Technologies: Oracle Corporation’s Vision.” Paper presented at Oracle Spatial
Special Interest Group Meeting, Denver, CO, March 18.
Louwerse, Abraham and Hans van DerMaarel. 2004. “Going Mobile.” Powerpoint
presentation. Wharton Conference.
MapServer. 2007. “Welcome to MapServer.” Minneapolis, Minnesota: University of
Minnesota. Available at http://mapserver.gis.umn.edu (May, 2007).
Marble, Duane. 2006. “Who Are We? Defining the Geospatial Workforce.” Geospa-
tial Solutions (May), 16(5): 14–21.
McFarlan, F.W. 1984. “Information Technology Changes the Way You Compete.”
Harvard Business Review, 62(3): 98–103.
Meeks, W. Lee, and Subhasish Dasgupta. 2005. “Geospatial Information Utility:
An Estimation of the Relevance of Geospatial Information to Users.” Decision
Support Systems 38:47–63.
Meeks, W. Lee, and Subhasish Dasgupta. 2005. The Value of Using GIS and Geospa-
tial Data to Support Organizational Decision Making, in In Pick, James B.
(ed.), Geographic Information Systems in Business, Hershey, PA: Idea Group,
pp. 175–197.
Murphy, L.D. (1996). “Competing in space: the strategic roles of geographic infor-
mation systems.” In Proceedings of the Association for Information Systems, AIS,
Atlanta, Georgia.
Oracle Corporation. 2005. “Oracle Grid Computing.” Oracle Business White Paper,
February, 12pp.
Oracle Corporation. 2005. “Oracle Spatial 10g.” Oracle White Paper, August, 7pp.
Available at www.oracle.com.
Pace, Scott, Gerald P. Frost, Irving Lachow, Dave Frelinger, Donna Fossum, Don
Wassem, Monica M. Pinto. 1995. The Global Positioning System: Assessing Na-
tional Policies. Santa Monica: Rand Corporation.
REFERENCES 37
Papp, Raymond. 2001. “Introduction to Strategic Alignment.” In Papp, Raymond

(ed.), Strategic Information Technology: Opportunities for Competitive Advan-
tage, pp. 1–24.
Pick, James B. 2004. “Geographic Information Systems: A Tutorial and Introduc-
tion.” Communications of the Association for Information Systems 14: 307–331.
Pick, James B. 2006. “Strategic Importance of Spatial Technologies and Enterprise-
wide Spatial Web Integration: A Case Study Analysis.” Paper Submitted to HICSS
2007 Conference.
Roush, Wade. 2005. “Killer Maps.” Technology Review 108(10): 54–60.
Silwa, Carol. 2003. “Even-driven Architecture Poised for Wide Adoption.” Comput-
erworld, May 12. Available at www.computerworld.com.
Smith, Gary, and Joshua Friedman. 2004. “3D GIS: A Technology Whose Time Has
Come.” Earth Observation Magazine, November.
Sonnen, David. 2004. “Spatial Information Management: The Road to Enterprise
Systems.” Powerpoint presentation. Framingham, MA: International Data Corpo-
ration.
Sonnen, David, and Henry Morris. 2005. “ESRI: Extending GIS to Enterprise Ap-
plications.” White Paper. February. Framingham, MA: International Data Corpo-
ration.
Sonnen, David, Henry D. Morris, and Dan Vesset. 2004. “Worldwide Spatial In-
formation Management 2005–2009: Forecast and 2004 Vendor Shares.” Market
Analysis. Framingham, MA: International Data Corporation.
Sonnen, David. 2006. “Pervasive Spatial Information Management.” Geospatial So-
lutions 16(1): 20–21.
Standard and Poors. 2006. “Oracle Corp.” New York: McGraw-Hill Companies.
The Internet Society. 2006. Data available at www.isoc.org, 6/06.
Tomlinson, Roger F. 2000. “An Overview: The Future of GIS.” ArcNews, Winter.
Tomlinson, Ralph. 2003. Thinking about GIS: Geographic Information System Plan-
ning for Managers. Redlands, CA: ESRI Press.
U.S. Census. 2006. “Data on Retail Trade Sales.” Available at www.census.gov. 6/06.
Wade, Tasha and Shelly Sommer. 2006. A to Z GIS. Redlands, CA: ESRI Press.
Williams, David. 2004. “The Strategic Implications of Wal-Mart’s RFID Mandate.”
Directions Magazine, July 29. Available at www.directionsmag.com.
CHAPTER 2
INFORMATION SYSTEMS AND GIS
INTRODUCTION
GIS has foundations in information systems, geography, and other disciplines.

In order to understand and apply concepts of Geo-Business, the basic concepts
of GIS and spatial analysis need to be explained. One of the starting points is
the foundation role of Information Systems in GIS. Although GIS grew up in
the public sector with prominent interest in the geographic side, today with the
growth of GIS in business, there is more perception of GIS as an information
system. Viewed as an IS, the concepts and principles developed in IS since
the 1950s become more significant, including architecture, security, systems
analysis and design, databases, knowledge building, enterprise systems, and
scalability (Laudon and Laudon, 2007). The chapter’s overview on infor-
mation systems first reviews some elements of the concepts of information
systems. Next the transformation model of information systems in business is
described. Today it is a broad view that encompasses the entire organization, a
theme that will be reinforced by considering enterprise systems in Chapter 4.
Information systems support users at different levels in the organizational
structure, from the low-level clerk up to the Chief Information Officer (CEO).
Current systems are becoming increasingly interorganizational, for instance
eBay’s web auction system whose users frequently represent organizations.
Another example is a firm’s supply chain software system, which interacts
with a dozen other companies’ systems for supply chain.

38
INTRODUCTION 39
The chapter turns to GIS concepts and definitions. This builds more
depth on the rudiments from Chapter 1, including spatial data, data capture,
layering, spatial analysis, spatial web services, and the underlying geometry. It
explains the process of moving from information with a locational reference,
say locations of trucks in a fleet, to associating the reference with the earth i.e.
through a coordinate system, and arranging the information in geometries on
map layers. Then the information can be analyzed by standard modeling as
well as spatial modeling and analysis, i.e. basing models on spatial locations
and relationships. An example is a model to determine how many trucks in
the fleet pass within 10 miles of major metropolitan areas of over one million
population. At the end of this process the results can be mapped so users
can visualize outcomes, summarized in tables, and/or fed into other business
systems for further processing.
A chapter section examines the essential referent disciplines for Geo-
Business, especially business, information systems, and geography, but also
statistics, economics, marketing, operations management, and environmental
studies. The human and organizational sides are very important including
psychology, behavior, and organizations. For instance, the business GIS an-
alyst shown in Figure 2.1 is using a GIS that is based on geography, IS, and
systems concepts, but his perceptions, comprehension, and ability to com-
municate and share results refers to the behavioral and organizational realm.
The connections are explained, so the knowledge base for business GIS can
be perceived more broadly.
Figure 2.1 GIS Specialist at Work on a Business Problem

40 INFORMATION SYSTEMS AND GIS
The case study of Baystate Health illustrates a medium-sized organization

with a dynamic leader in GIS. It shows how GIS finds a place in a technology-
laden environment and reinforces information systems, organizational, and
GIS concepts.
Spatial data differ from some traditional business data. For instance, ac-
counting data are include in balance sheets, income statements, and return
on investments. This information is missing spatial-referencing. The section
on spatial data starts by examining how spatial-referencing takes place. It
covers essential rudiments of data capture, and coordinate systems. It carries
the concept of spatial data further and examines how they are incorporated
into data structures and databases.
The chapter then turns to elements of spatial analysis and spatial modeling.
In analysis and modeling, spatial information is examined for its relationships,
agglomerations, proximities, and trends.
OVERVIEW OF INFORMATION SYSTEMS
Information systems function in businesses to solve problems and help busi-

ness decision makers. They have become critical at all levels in the organiza-
tion. They are spreading through businesses, with increasing percentages of
business workforces using them.
A broad model of information systems for business problem solving, seen
in Figure 2.2, includes a range of entities. The diagram indicates the influence
of each of these entities for IS systems to function well. People are essential
to IS. They plan, develop, manage, and use information systems. They are
also the decision-makers that IS supports. Without people, IS in business
Business
Problem
Business
Solution
Information
System
People Organization Data Technology
Figure 2.2 Information Systems Model for Business Problem Solving

OVERVIEW OF INFORMATION SYSTEMS 41
would accomplish little. The organization is the broader social, behavioral,

and strategic setting for the information system. For one thing, most of the
people, technology, and data for the system are housed within the organization.
Its size, structure, leadership, missions, and strategies influence how, why, and
where the information system functions. The goals in the ultimate IS solution
are usually to help the organization achieve better performance, productivity,
or value.
Data from inside and outside the organization are input into the IS for pro-
cessing. For instance, on the inside are accounting, financial, marketing, sales,
human resources, legal, manufacturing, inventory, R&D, and administrative
data. From outside the organization are economic, marketing, political, envi-
ronmental, urban, technological, consumer-behavior, and demographic data.
Data from the outside derive from governments, consulting firms, survey
organizations, media, competitors to the extent possible, and other sources.
The data are increasingly available online, allowing access from different
locations globally.
The technology is critical to any information system. It includes computer
hardware, communications networks, sensors, satellites, mobile devices, and
data storage. It is assembled together to function as an integrated and reli-
able technology base. Today, some or all of the technology is increasingly
located at an outsourcer, web service provider, communications carrier, or
partner company. The rise of outsourcing is due to globalization and cost
savings. With the internet and lower communications costs, some services
dependent on technology can be moved to lower-cost settings including
overseas.
A factor ever present with the technology is the increase in processing
and storage capacity. Moore’s Law predicts that processor capacity will
double about every two years. For instance in Table 1.2, it was seen that
processing capacity grew by about half a million times between 1971 and
2006, which is doubling every two and two-thirds years (Intel, 2007). Ex-
perts predict a continuing rapid rate of capacity increase (Hiremane, 2005).
This parallels other rapid and continuing increases in technological capaci-
ties and throughputs. For instance in the communications realm, broadband
bandwidths defined by the ITU as 1.54 million bits per second or higher
apply to increasing number of business users and to the general population.
For the general public in December of 2003, 70.5 percent of South Ko-
rean households have broadband, while the figures for Hong Kong (50.3
percent), Canada (36.2 percent), U.S. (22.5 percent), and Belgium (24.7
percent) are also high (Austin and Bradley, 2005). Measured slightly dif-
ferently, by the percent of total population that had broadband, the results in
2006 were: Denmark (31.9 percent), Netherlands (31.8 percent), Switzerland
(28.5 percent), Canada (23.8 percent), Japan (20.2 percent), the United States
(19.6 percent), and Australia (19.2 percent) (OECD, 2007). These propor-
tions can be expected to grow, with broadband becoming commonplace in
economically advanced nations. Along with higher capacities, is a much wider

acceptance and use of web-based technologies, based on the infrastructure to
support it.
Another influential technological change is mobile technology. Cell phones
increased to several billion worldwide and an estimated 200 million in the
United States. Mobile communications are widely utilized in advanced coun-
tries and growing rapidly in developing ones. While the majority of older
phones have limited data capability, new 3G services provide much greater
data bandwidths, offering the potential for new applications. Besides cell
phones, prominent mobile technologies include pocket pcs, personal device
assistants (PDAs), GPS units, RFID equipment, iPod family of devices, med-
ical monitoring devices, and many more.
Thus the technology element of IS continues to expand dynamically. In
Figure 2.2, these four key elements discussed influence the information sys-
tem. An information system collects, processes, stores, analyzes, and outputs
data and information for a specific purpose (Laudon and Laudon, 2007). The
type of system developed and built depends on the type of business problem,
the decisions being made, the users, and the organization.
The output from the IS is intended to support a business problem as seen
in the figure. If successful, the business problem will be solved and a business
solution will be reached. If not successful in solving the problem, the system
can be re-run, or other ways to solve the problem attempted. IS is only one of
many tools that are used in business problem solving For instance, a customer
is being tracked by a sales team using a customer relationship management
(CRM) system. The data are accurate and the technology base works well. The
system runs well and provides the sales force with some relevant information
to help land the customer. However, if the customer decides to do business
with another company, the problem is not solved. The point is that the goal of
an information system is to provide support in the decision making process.
However, there is no guarantee that the problem will be solved. Whatever the
outcome, whether the problem is solved or not, the results are fed back to
influence the IS in the future.
Having set this broad context for information systems, two other rudi-
mentary concepts of IS are essential, the systems development process, and
the process of data transformation into information and knowledge. The sys-
tems development process, sometimes referred to as systems analysis and
design, supports the building of new information systems and the upgrad-
ing of existing ones. Every system currently in place and operating went
through the systems development processes. The standard steps in developing
a system are: planning, analysis, design, implementation, and maintenance
(Laudon and Laudon, 2007; Satzinger et al., 2007). For instance, a small
business is very disappointed with its scheduling system. Consequently in
the field, its marketing people cannot keep track of their daily appointments
with customers. The company gets a rough idea of some solutions, studies
GIS: CONCEPTS AND DEFINITIONS 43
competitors, determines what the cost ranges may be and gets the buy-off
from the company president (planning). It then looks in depth at alterna-
tive systems, checks with users, does careful comparisons of alternatives,
and recommends the solution to management (analysis). The current sys-
tem is considered as one of the alternatives. In design, the new system
is carefully analyzed and specified in detail including user needs, inputs,
outputs, data, data flows, process flows, and workflows Some features of
the current system might be considered. In implementation, the new sys-
tem is either built in-house or outsourced to an external provider to build.
It is tested; users are trained; and it is turned on for use. The process
then moves into an extended step of ongoing use of the system, often for
many years (maintenance). There may need to be bug fixes, changes, re-
pairs, and further training. At some point the system will age or deteriorate
to the point that it is either replaced by a new one or undergoes a major
upgrade.
Systems Development for GIS is the focus of Chapter 6. It is fortunate that
the IS field has developed not only the concept of the stages just mentioned, but
methods, techniques, and management approaches to systems development
that carry over to GIS.
The final IS concept in this section is the process of data transformation to
information and knowledge. Data are defined as raw facts that do not have
context. For instance, this might be a set of thirty numbers or thirty cars.
Neither of these has much value beyond the immediate facts. Information
is fact that are organized so they have value beyond just being known. For
instance, if the cars have been sorted into four brand groups, there is addi-
tional value and they become information (Haag et al., 2005). Knowledge is
information that has a higher level of interpretation through being applied in
a certain context. It requires business intelligence, models, and/or experience
to have the increased value from information. For instance, the cars have now
been tested for performance and the dealership has grouped them into per-
formance quality categories. The added experience and greater context turns
the information into knowledge. In Figure 2.2, data become information and
knowledge as part of the processes through one or multiple cycles. After
several cycles, the box labeled “data” actually becomes a mixture of “data,
information, and knowledge.” In the chapters upcoming, the business prob-
lems and decisions in cases and examples involve at times all three of these
concepts.
GIS: CONCEPTS AND DEFINITIONS
This section expands on Chapter 1’s initial introduction to GIS and spatial
concepts. GIS both builds on the structure of IS concepts as well as on new
ones that are particular to space and geography.
The major principles discussed for Information Systems hold true for
GIS, which constitutes one type of information system. Referring again to
Figure 2.2, a Geographic Information System might be present in the place
of an Information System. For GIS, the same entities are present of people,
organization, data, and technology. However, the influences are more specific
in the following ways:
People. For a GIS, the people need to include some staff and/or outside con-
sultants who are knowledgeable about GIS and spatial concepts. This might be
GIS specialists with training and experience concentrated in GIS. It may be IS
specialists, networking, and other technology people, who have received some
training in GIS and serve partly as GIS staff. Or it might a skilled GIS and
spatial group at an outsourcer. For larger companies, it is likely a combination
of several of these.
Another set of people that need to be present are non-specialists in the organi-
zation, such as executives, managers, people in key GIS user departments, and
other support units, who have enough exposure to GIS and spatial technologies
to appreciate what they are and why they are useful to the organization. If such
a group of GIS-aware people is not present, then awareness training might be
needed.
The internal users, users in partner organizations, and customer-users of the
system also need to be exposed to GIS and spatial technologies through training
and communications. If the spatial application is more sophisticated to use, they
would need more of this. For instance, if a multi-featured GIS application is
implemented for a large building construction firm, the users including GIS
analysts, data-base administrator, systems support people, and programmers,
would need more training.
Organization. The organization for a GIS application has all of the ordinary
IS roles that have been described. For GIS and spatial applications, the or-
ganization needs to give added emphasis to several aspects. First, since GIS
is more recent in business, it needs to emphasize organizational learning and
training. Second, it needs to provide sufficient support to the GIS special-
ist function described under People. GIS is often not as established in busi-
ness as IS, which has been around for a longer time, so the GIS department
or group needs sufficient resources to thrive. For instance, for two of the
book’s twenty research case studies, the GIS manager mentioned in the in-
terview that he/she had not received adequate support from the organization
and in one case reported that the IS department resisted GIS including ma-
neuvering to reduce their resources. However, the good news is that mostly
GIS was given organizational support and had a good working relationship
with IS.
A third aspect for the organization is to gain understanding that GIS develop-
ment cycle times may be long. This point is emphasized in Chapter 7. The rea-
sons development duration may be lengthy include (1) the necessity to identify
and gather spatial data, check it for errors, and often transform it for use, (2) the
added time for spatial learning and training, (3) the coordination time between
GIS and other associated departments such as IS, networking/communications,
engineering, and marketing, and (4) the need sometimes to associate several
spatial technologies together, such as server-based GIS, GPS, mobile devices,

and sensors.
In short, the organization has all of the ordinary functions for IS plus a number
of crucial ones more specific to GIS. This includes dealing with often large
amounts of spatial data. In addition, the organization is often challenged by the
novelty and rapid innovation pace of spatial technologies.
Data. The data part of the overall scheme has additional aspects with GIS and
spatial technologies of locating, cleaning, transforming, and inputting spatial
data—in addition to the normal processes to handle attribute data. Even though
today many spatial datasets are available through the internet and web, prob-
lems remain. Those datasets may not be appropriate to the business problem.
They may be restricted from use through copyright or licensing, or may be
in a proprietary format that prevents use. They may be of low data quality,
a topic covered in Chapter 8. Further, such a dataset may need to be trans-
formed, in order to correspond to spatial data from other sources being used
in the same GIS. In addition, for some business problems, the best spatial
data may need to be newly collected—they don’t already exist. Data can be
collected through means of digitizing, geocoding, GPS, scanning aerial pho-
tos, and conducting surveys that are discussed in the section on understanding
spatial data.
Technology. As emphasized in the first chapter, GIS and spatial technologies
have dynamically changed during their over forty years of commercial use,
and they continue to evolve. In even the past ten years, among the advances
are: mainstream commercial use of RFID, consumer-web spatial services, spa-
tial data warehouses, specialized spatial mobile devices, and geospatial vir-
tual reality. IS technology has also advanced at a rapid pace. However, the
specialized knowledge and capacity to plan for emerging spatial technolo-
gies needs to be present but often is reduced or missing in a traditional IS
department.
Another special aspect of GIS and spatial technologies is the need to inte-
grate them. They need to be integrated with each other, and with enterprise
and other mainstream systems. This can sometimes be challenging. For in-
stance, some leading enterprise systems such as ERP do not yet have built-in
spatial capacity. Instead, there are specialized “connectors” between them and
GIS software. However, often the “connectors” may be difficult to imple-
ment and operate, as will be seen in Chapter 4. In sum, the inclusion of GIS
brings additional challenges to IT management, particularly the specialized
need to track GIS’s rapid advances and often heightened system integration
challenges.
Most of the remainder of this chapter focuses on details of how GIS is

constituted and how it functions. It will cover many of GIS’s unique aspects,
including coordinate systems, spatial layers, connection of layers to tables,
spatial analysis, capturing of spatial data, spatial models, and GIS software.
One foundation IS concept that carries over very well in its principles to
GIS is systems development. The five major steps outlined for IS apply to
GIS and will form a basis for the coverage of spatial systems development in
Chapter 6.
The discussion now turns to rudiments of GIS functioning. It starts with

GIS location and coordinates, map layers and their relationship to attribute
tables, and discusses what is spatial analysis and provides an example. This
builds on the starting material in Chapter 1.
GIS Location and Coordinates

A GIS functions by the input of information that has a location. This is
commonly done by identifying the X-Y earth coordinates of an object at
a time and place. The most common coordinate system is longitude and
latitude. A longitude is the east-west angle of a point starting at 0 degrees for
the Greenwich, England, observatory. As seen in the figure, they are calculated
by determining what meridian an earth point lies on. A meridian is a line on
the surface of the earth extending from the North Pole to the South Pole,
always at the same longitude. For instance, any point lying on north-south
meridian that is at a 22.5◦ angle to the West of Greenwich, England, lies on
the meridian of −22.5 degrees longitude.
The latitude is defined as the angle created by a vertex at the earth’s
geometric center and lines to the equator and a point on the earth’s surface.This
angle for the north pole is +90 degrees and for the south pole is −90 degrees.
Lines circling the earth at a latitude are called parallels, for instance the circle
at 45 degrees north of the equator is the 45th parallel (see Figure 2.3). This
leads to any point of the earth’s surface be identified as (X,Y), where X is the
latitude and Y is the longitude.
Figure 2.3 Concepts of Longitude and Latitude. Source: Bolstad, 2005

Figure 2.4 3-D Visualization of a Park in London, England, Using ArcInfo. Source:
ESRI
A third dimension can be added for elevation, resulting in (X,Y,Z) co-

ordinates. For GIS, three coordinates allow 3-D displays and analysis. For
instance, a 3-D display of a park in London, England (Figure 2.4), shows
details of the park layout and visualization of buildings surrounding the park,
although more the distant buildings have simpler shapes. Such a display could
be useful for architects, builders, and city planners. The calculations and pro-
cessing of 3-D GIS are more intensive than 2-D and special algorithms and
solutions are required (Smith and Friedman, 2004). Software vendors have
released special software to support it, such as ESRI’s ArcView 3-D Analyst
(ESRI, 2007).
For maps in (X,Y) coordinate systems, how does the recording of a point
location translate into a complex map? This is accomplished by building up
two or more points into lines and polygons (i.e. areas). As seen in Figure 2.5,
on the left side are a set of seven (X,Y) coordinates in the form of two numbers.
The numbers represent the storage of points in a computer. The points are
shown by the black points. A pair of points in turn forms the ends of a line.
A group of points can form a connected set of lines, as seen to the left of the
points. Finally a connected set of lines can form the boundary of a polygon. If
the line segments are close enough together, the polygon will appear to curve
smoothly. Having formed points, lines, and polygons (areas), they in turn can
be combined in a set of layers to portray complex maps. In short, the (X,Y)
point locations form the basis for constructing eventually any map shape.
Figure 2.5 Points, Lines, and Polygons
Map Layers and Their Relationship to Attribute Tables

Once points, lines, and polygons of interest are identified, they can be grouped
into map layers.
As introduced in Chapter 1, a GIS starts by the input of attribute and
spatial data. The attribute information is mostly numbers and characters; the
spatial data are (X,Y) coordinates of points. The GIS functions by combining
attribute data in attribute tables with corresponding spatial data in map layers.
This was introduced in Chapter 1 as the generic GIS design shown in Figure
1.2. On the right side of the model in Figure 1.2, each map layer represents a
map feature. The information in one layer can be in the form of either points,
lines, or polygons.
For instance, one map layer consists of the points for a company’s thirty-
eight distribution centers. This layer would be added to the right in a diagram
like Figure 1.2. On the left side of the model are the associated tables. In the
example, the table might contain four attributes for each of the thirty-eight
distribution centers. The attributes are: floor area (in square feet), number of
items received per month, number of items shipped per month, and number
of employees in a center. The table has thirty-eight rows (one for each center)
and four columns (one for each attribute).
This is made clearer in Figure 2.6, that shows three layers in a GIS, the top
one consists of seven polygons (areas) and an associated table having three
attributes. The second layer consists of four points and has an associated table
with four attributes, while the third layer of five lines has an associated table
with two attributes. For the polygon layer, each polygon has its associated
attributes in a row of the table. For instance Polygon 2’s attributes A, B, and
C appear in the row 2 of the table. The number 2 is referred to as the feature
ID. It appears as the first column in the table, which serves to connect the
layer with the table.
The same process applies for the points and the lines. For instance, Line 23
is associated with the attributes P and Q in the third row of the table, having
Feature ID 23.
Spatial Data Attribute Data

Attribute A Attribute B Attribute C
1 2 1
2
3
3 4 5 4
5
7 6
7
6
WW
Attribute G Attribute H Attribute I Attribute J
XX WW
XX
YY
YY ZZ
ZZ
21
Attribute P Attribute Q
22 21
22
23 23
24 24
25
25
Figure 2.6 Relationship of Spatial Data and Attribute Data. Source: Adapted from
West, 2000
This simple example can be extended to a larger GIS that may have thirty
layers with dozens of attributes per layer. The same principle holds that for a
layer, each spatial feature has a unique feature ID that also connotes the row
of the table that contains its attributes.
Spatial Analysis—What It Is and an Example
A GIS functions by performing spatial analysis on the information stored in
map layers and their associated tables. A simple business example is now
presented on how the spatial and attribute data inform spatial analysis.
The business problem is to analyze spatial relationships involving a fast-
food company XYZ Inc. in an urban area that has four sales districts and a
network of major highways that are located nearby XYZ’s fast food outlets.
As seen in Figure 2.7, the GIS has two layers. The bottom layer contains
the city’s major highways. Associated with each major highway segment
(represented by straight line, or curved line segments) are four attributes:
number of gas stations segment, number of gas stations that sell soft drinks,
the highway speed limit, and the total daily vehicle traffic for the segment.
The second layer consists of four areas (polygons), each representing a
sales district for XYZ. Associated with each sales district are five attributes:
monthly sales, number of employees, customer floor area, number of menu
items, and soft drinks as percentage of sales.
Analysis of Length of Highway Per District, Compared

to Monthly Fast Food Sales Per District.
Analysis of Number of Gas Stations with Mini-Markets

per District, Compared to the Area of Districts
Attribute
Tables Map Layers
Monthly Sales
No. of Employees
Customer Floor Area B
A
No. of Menu Items C
Soft Drink Percent of D
Fast Food
Sales
Sales
Districts
Gas Stations on Highway

Gas Stations with Mini-
Markets on Highway
Highway Speed Limit
Vehicle Traffic Daily Highway
Transportation
Network
Figure 2.7 Fast Food Sales Regions and Transportation Network: Example of
Spatial Analysis
In this example, spatial analysis for XYZ is shown in the box at the top
of the figure. The first analysis compares, by sales district, the length of
highway in the district versus the average monthly fast food sales in the
district. When the highway map is superimposed over the sales district map,
it is possible to determine the total length of highway segments within each
of the four sales districts. This is referred to as spatial analysis because
spatial features, not attribute features, are being compared and analyzed to
determine an outcome. After the length of highway is computed within each
district, it is divided by the average daily food sales per district. This yields
the highway length per sales total, so that the districts can be compared
to see if some districts are not realizing enough sales given the amount of
highway.
The second analysis evaluates how much the presence of gas stations with
mini-markets compares to the area of districts. XYZ would like to know if
larger districts have more gas-station mini-markets. It consists of calculating,
for sales district, the ratio gas stations with mini-markets to area. If the ratio
LINKAGES TO OTHER DISCIPLINES 51
is high, then XYZ’s marketing department will give more emphasis to gas
stations. It is spatial analysis because the GIS needs to compute the overlap
of highway segments and sales districts.
This example demonstrates that even in a simple setting, spatial analysis
can yield new insights into business problems. The insights are in addition
to other standard analysis and models that could be run using just the at-
tribute information i.e. non-spatial analysis. The chapter now returns to the
broader perspective on GIS and addresses how it fits into academic disciplines.
Baystate
LINKAGES TO OTHER DISCIPLINES
GIS began in the 1960s and 1970s in departments of geography, landscape

architecture, cartography, and remote sensing (Longley et al., 2000a). Over
the past two decades, it has branched into other academic disciplines partic-
ularly computer science (Longley et al., 2000a); statistics and geostatistics
(Getis, 2000); land administration (Dale and McLaren, 2000); urban plan-
ning, regional sciences, public policy (Greene, 2000); natural resource, en-
ergy, and environmental studies (Harder, 1999); social sciences, medicine,
and public health (Khan, 2003); and the humanities (Gregory et al., 2002).
Since the 1990s, it has spilled over into the disciplines of business includ-
ing management (Huxhold, 1991, 1995; Tomlinson, 2003), information sys-
tems (Huxhold, 1995), logistics, organizational studies (Reeve and Petch,
1999), real estate (Castle III, 1998; Thrall, 2002), retail management (Long-
ley et al., 2003), marketing (Miller, 2007), and telecommunications (Godin,
2001).
These disciplines have all made conceptual contributions to GIS for busi-
ness. As shown in Table 2.1, the concepts contributed are essential ones
for this book including spatial analysis, systems analysis and design, deci-
sion support systems, visualization, geodemographics, cost-benefit analysis,
strategic management, and web service models. It will be seen in the book
that some of the key literature that supports business GIS comes from these
other disciplines.
The blending reveals that GIS today goes across the spectrum of knowl-
edge and relates to a wide variety of disciplines. This is reflected by some
schools and universities that have elevated GIS to school-wide or campus-
wide focus. The blending of knowledge also has implications for businesses.
They also consist of a variety of departments drawing on distinctive knowl-
edge bases. It means that in a business, people from marketing, IS, fi-
nance, and real estate have valuable ideas to contribute since their knowl-
edge is part of the broader GIS and spatial knowledge. GIS problem-solving
teams in a business can benefit by inclusion of people from several referent
areas.
c02
TABLE 2.1 Referent Disciplines for Concepts and Theories of GIS

Concept of Theory in GIS in Concept or Theory in GIS in

Business Referent Discipline(s) Business Referent Discipline(s)
Spatial Analysis Geography, Regional Science Networking Configuration Telecommunications

Location Theory Geography, Operations Research Visualization Psychology, Computer Science,
Media Studies, Geography,
Earth Sciences
October 18, 2007
Gravity Model Geography Geostatistics Statistics

Remote Sensing Geography, Earth Sciences Customer Relationship Marketing, Information Systems
Management
13:23
Decision Support Systems Information Systems Adoption/Diffusion Theory Marketing
52
Knowledge-Based Discovery Information Systems Market Segmentation Marketing, Retail Management
GeoDemographics
Data Mining Information Systems CAMA and AVM Models Real Estate
Location Based Services Information Systems Cost-Benefit Analysis Economics, Business
Char Count=
Value of II Investment Information Systems, Economics Organizational Theory Strategic Management, Sociology
Management
Electronic Business Information Systems, Economics End User Computing Systems Information Systems
Analysis and Deisgn
Environmental and Geologic Natural Resources and Data Models, Web Service Computer Science
Time-Based Models Environmental Studies Structures
Epidemiological Spatial Models Medicine and Public Health Mapping with Incomplete Data Humanities
CASE STUDY: BAYSTATE HEALTH 53
CASE STUDY: BAYSTATE HEALTH
Baystate Health is a very large health-care system in New England. It is based

in Springfield, Massachusetts, and has three member hospitals in central and
western Massachusetts. There is a children’s hospital and a variety of facilities
for health services as well as health centers. GIS is part of the health geograph-
ics program that has been in place since the mid 1990s. The GIS applications
started in the Department of Surgery. The department, an accomplished prac-
ticing surgeon, came to Baystate with responsibility to re-energize Baystate’s
research program. There the first Baystate GIS application was identified and
implemented in injury prevention. A biostatistician was added who had a GIS
certificate, followed by two GIS professionals. There is now a four-person
department chaired by the chief surgeon, which makes this department
unique among U.S. hospitals in its size and experience (Kolb, 2001).
GIS is utilized in a wide variety of applications within the Baystate system.
They include:
Emergency planning. Baystate participates in multi-state regional emergency

planning. In multi-state emergency exercises, Baystate has provided situational
maps and other GIS support to the Regional Coordination Center.
Epidemiology. Epidemiology has been a long-time part of the GIS at Baystate,
beginning as part the chief surgeon’s emphasis on research. Early on, GIS
helped in epidemiological and statistical analysis of breast cancer and injuries,
and it extended to youth violence. Currently Baystate is collaborating with
United Way in western Massachusetts to produce a CD of epidemiology to
distribute to all health services in that part of the state.
Routing. GPS-enabled route optimization and route management are avail-
able for two delivery systems—homebound patients and lab specimens. Some
in-house software additions have been made to commercial software.
Hospital floor plan maps. A disaster patient tracking system links the master
patient database to floor plan maps of the hospital, which would support the
hospital not just in finding a bed but in finding a bed located near the appropriate
services.
Emergency services. Baystate’s emergency services center utilizes GIS for
dispatch and to study spatial patterns of former emergency patients, examin-
ing the services they received compared to their demographic and financial
attributes. It has the goal to identify patients who should not be going to emer-
gency rooms.
Human body. Baystate has designed and implemented a unique GIS for view-
ing the human body in 2-D and 3-D. It has been notably applied to understand
the locations and statistically analyze the locations of colon polyps and lesions,
including cancers, based on a new coordinate system for this study.
In accomplishing this wide set of applications, the Health Geographics

team has collaborated with other Baystate departments and multiple external
government and nonprofit organizations—in emergency planning, commu-
nity health care, and epidemiology. Led by a surgeon-champion of GIS use,
the team has acted as a change agent for the whole organization, moving it
into a national leadership position in hospital GIS. Curiously, almost none
of Baystate’s administrative systems have so far been spatially-enabled, with
the exception of some aspects of facilities management (floor plans). GIS
has stayed on the medical and health side. This may be why GIS is not yet
recognized as strategic by Baystate’s top management, many of whom may
be closer to the administrative and financial side of Baystate.
The case illustrates many of the chapter concepts. It demonstrates the
influences seen and discussed for the GIS version of the model in Figure 2.2.
It confirms the importance of people and the organization in achieving GIS
solutions. Leadership is seen to have been essential for the build-up in only ten
or so years to national prominence. The case demonstrates the significance of
an interdisciplinary body of knowledge of GIS. At Baystate GIS benefited by
people from diverse health-care departments being involved. There remains
the potential to have the same impact on the business/administrative side of
the organization.
UNDERSTANDING SPATIAL DATA AND MAPPING
This section examines spatial data and mapping. These are large topics that
readers can find a lot more about in sources that focus fully on them (Brewer,
2005; Bolstad, 2005; Clarke, 2003; Heywood, 2006; Longley et al., 2006b).
Spatial data can come ready-made from governments, businesses, con-
sultants, nonprofits, and portals. Among the prominent locations to ob-
tain pre-set spatial data are the GeoData.gov (sometimes called Geospatial
One-Stop), U.S. Census (www.census.gov), and The Geography Network
(www.geographynetwork.com). GeoData.gov can help a user locate a data or
map service, create a map, browse information regarding geospatial data, and
cooperate on data acquisitions. The latter feature is a bulletin board for locat-
ing partners or seeking partners in acquiring spatial data. It’s possible to use
the “Marketplace Map” to indicate a certain geography is of interest for part-
nering, and a list will be produced of prospective partners for that geographic
area. There is a grants area to find opportunities for geospatial funding. A
sample map created by GeoData.gov’s Map Viewer (see Figure 2.8) shows
the transportation network encompassing Kansas City and Wichita, Kansas.
The U.S. Census site has a wealth of economic, demographic, and social
information and maps for the United States. The maps extend down to small
areas within cities and cover hundreds of variables. The user can create
maps using the Census map service tools, or download data to use with a
commercial software package. The Geography Network is a nonprofit spatial
data portal that provides maps and data, both free and at cost, for geographies
worldwide on subjects covering a vast range of topics, including geography,
geology, environmental, social, and business. Some of the data areas are
richer with data than others, for instance a wide range of environmental
UNDERSTANDING SPATIAL DATA AND MAPPING 55
Figure 2.8 GeoData.gov Map Viewer Display of Transportation Network that In-
cludes Kansas City and Wichita, Kansas. Source: GeoData.gov
offerings are provided, but the business ones are limited, likely due to the
proprietary nature of much business data. The portal is sponsored by ESRI
Inc. and supported by Sun for hardware. It allows sharing among private and
governmental providers of data, web services, and other users globally.
In attempting to locate appropriate existing resources, “the trick is knowing
where to look, what to do when you find what you want, and how to get the
data into your GIS” (Clarke, 2003).
If commercial map sources do not provide an appropriate map or spatial
data, then other means must be used to capture spatial data. Among the major
ways are digitizing, scanning, using GPS devices, or utilizing satellite and
remote-sensed data.
Digitizing
Digitizing manually transcribes an existing hardcopy map into digital form.
As seen in Figure 2.9, a user moves a digitizing puck across the hardcopy map
fastened to a digitizing tablet and follows click sequences to enter the points
with (X,Y) coordinates. Digitizing is time-consuming and error-prone but
might be necessary today if only the hardcopy map is available as a source.
Figure 2.9 Digitizing Mouse and Tablet. Source: © GTCO CalComp, Inc., used
by permission
Scanning
A hardcopy map can be entered using a scanner. After entry, software pro-
grams along with a mouse can be used to identify features on the screen, i.e.
points, lines, and polygons. It is quick to scan a map, but the identification of
features remains manual and time-consuming.
Use of GPS
GPS provides very accurate positioning by connecting to a system of twenty-
four satellites that orbit the earth at middle range of about 12,000 miles.
The GPS receiver receives signals from those satellites that are not blocked
by the earth. Normal GPS accuracy is within a ten to twenty-five meter
range, but when used differentially, i.e. with one device fixed and another
moving around, the range can be sub-meter. An example of a GPS unit that
is hardened for field use is seen in Figure 2.10. The GPS can read at timed
UNDERSTANDING SPATIAL DATA AND MAPPING 57
Figure 2.10 Hardened Trimble GPS Device

intervals or when the user requests. The (X,Y) location is stored in the unit,
and depending on its design can be transmitted immediately or on return to
a fixed workstation. GPS has become prevalent today in some countries such
as the U.S. that have mandated that new phones need to to be able to identify
their location precisely, predominantly by GPS capability.
Satellite and Remote Sensing

Satellite imagery became available commercially starting in 1972 with the
Landsat satellites. Today such services as SPOT, INTELSAT, and the NOAA
satellites provide extensive satellite imagery worldwide. Examples of satellite
and aerial photography images at different scales and resolutions are seen in
Figure 2.11 (Clarke, 2003). This provides exact land features that are often
superimposed on feature maps, so both are available. The accuracy is high,
but locating the correct imagery with the right resolution and type of image
takes training and experience. Also, satellites are limited by land cover, so
one cannot see beneath nature’s canopies or into built structures.
Once spatial data have been entered into a GIS data-base, they are expressed
in layers and processed in ways that have been discussed. The resulting maps
can be displayed in hardcopy, on monitor screens, and in mobile devices.
They can be displayed along with tables and other imagery such as photos.
Figures 2.12 and 2.13 show several examples of map display. A handheld
street map of part of the city of Redlands, California, is seen in Figure 2.12,
while a digital photo of a feature and the corresponding satellite map appear
in a screen display in Figure 2.13.
GIS ANALYSIS
GIS analysis techniques consist of methods that are used in the spatial anal-
ysis, modeling, and statistical analysis of a GIS. This section describes a few
of the most common spatial analysis techniques. A full treatment is available
in specialized sources (Mitchell, 1999; Getis, 1999; Lo and Yeung, 2002;
Longley and Batty, 2003). Spatial analysis consists of analytical techniques
that emphasize the map layers, portrayed on the right side of Figure 1.2. It
relates and compares the features of the physical locations of objects in space
(Getis, 1999; Longley and Batty, 2003). Since it draws from geography, it is
not familiar to most IS researchers. Modeling and statistical analysis methods
include many methods and techniques well known in business disciplines, but
often modified to take into account spatial relationships. These methods are
based on attribute and spatial data, portrayed on both sides of Figure 1.2. Spe-
cialized statistical methods that include space are referred to as geostatistics
(Getis, 1999). Some desktop/laptop software packages include geostatistics
modules, such as ArcGIS Geostatistical Analyst (Johnston, 2001).
GIS ANALYSIS 59
Figure 2.11 Imagery Used in GIS, (a) Landsat image, (b) Digital Orthophoto of
Washington DC, (c) Aerial photography of San Francisco with vector GIS overlay.
Source: (a) and (b) Courtesy of USGS, (c) City and County of San Francisco
Spatial Analysis
Since spatial analysis techniques compare spatial features, they can, for ex-
ample, determine how many points are inside a polygon, how many line
segments cross a polygon boundary, or how much polygons overlap each
other. A practical example would be to ask how many highways and streets
(lines) cross a zone of retail businesses.
Figure 2.12 Handheld GPS Unit Showing Redlands Streets and Businesses
Figure 2.13 Digital Photo and Associated Satellite Image Displayed Together
GIS ANALYSIS 61
Buffer Analysis
In buffering, GIS software forms bands on all sides of a point, line, or polygon,
in order to perform analysis within the bands. A simple example would be
to assign half-mile buffers on both sides of a highway, in order to query how
many service stations are within the buffer.
Map Overlay Functions

Overlay consists of superimposing two or more map layers. Quite a few
operations for map overlays are typically supported. One example is the
union procedure, in which two polygon layers are combined to result in a
third output polygon layer. The resultant layer has the identical geographic
extent as the two layers that form it, and it concatenates the attributes from
those two layers in its associated data-base (Greene and Stager, 2005). The
polygons from the original two layers are cut up wherever they cross to form
a larger set of polygons in the output layer as seen in Figure 2.14.
This diagram shows one layer that has census blocks and a second layer
with three store sales areas. The resultant overlay on the bottom combines
all these features. The attributes of each layer are combined into a resultant
table (not shown). The advantages of overlay procedures include the ability
for form new layers that can provide new insights. The business purpose in
the example is to understand the complex pattern of census block attributes
and how it relates to the sales areas (Greene and Stager, 2005).
Examples of more sophisticated spatial analyses include uses for industrial
specialization and location quotient analysis in an urban labor market and for
trade area analysis, based on a gravity model, which can examine customers’
attractions to competing consumer destinations. Greene and Stager (2005)
discuss the gravity model with respect to attraction of major opera houses in
northern Illinois and southern Wisconsin.
Proximity Analysis
Proximity analysis assesses how close certain map features are to other map
features. For instance, it can determine how many customers are in a grocery
store’s trade area.
Figure 2.14 Map Overlay Example. Source: Modified from Greene and Stager,
2005
Longitudinal Change Detection

This type of analysis compares maps over time, and assesses what significant
spatial changes have taken place. For instance, real estate maps produced for
a suburb on an annual basis can show variations in the spatial patterns of
housing prices and their changes over time.
Modeling and Forecasting

Forecasting and simulation models can be built with spatial data, and the
results displayed in map form. An example is a model that projects the
population distribution of retail consumers within a county, based on starting
year data. The future population distributions of retail consumers can be
mapped, to inform businesses and the public.
Statistical Analysis
Statistical models are often applied to study the relationships of certain spa-
tially referenced attributes to other attributes (Getis, 1999). They include
correlation, regression, analysis of variance (ANOVA), cluster analysis, prin-
ciples components, and t tests. Often the methods are corrected or refined
for geostatistics (Getis, 1999). The input data, as well as the results, of many
of these models can be represented as spatial displays. This enhances un-
derstanding of the geographical effects and influences. Although beyond the
scope of this chapter, a specialized part of statistics, called geostatistics, takes
especially into account spatial effects and interactions, such as spatial auto-
correlation (Getis, 2000). An example is a regression forecast of the future
spatial patterns of demand for advertising services in an urbanized area.
EXAMPLE: LOCATION QUOTIENT ANALYSIS FOR SITING
A common problem facing businesses is the decision on where to site new

facilities that include retail outlets, distribution sites, factories, service offices,
and administrative offices. The decisions are often influenced by concentra-
tions of business facilities for the industry as a whole. For example, profes-
sional services firms often group together in the same urban zones to share
workforce pools, services, and customers. Similar agglomeration is present
for locational groups of hotels and restaurants. The location quotient (LQ)
indicates the ratio of proportion of workforce in a particular industrial sector
for an urban zone to the proportion of workforce in that industrial sector for
the entire urban area. An LQ of 100 indicates the proportion of workforce
in the zone is the same as that for the urban area, while an LQ of 400 indicates
the zone’s proportion is four times the urban area’s (Greene and Stager, 2005;
Greene and Pick, 2006).
EXAMPLE: LOCATION QUOTIENT ANALYSIS FOR SITING 63
Figure 2.15 Spatial Analysis of Industrial Locations for Los Angeles Using Loca-
tion Quotient. Source: Greene and Stager, 2005
LQ spatial analysis maps the levels of LQs for an urban area. For exam-
ple Figure 2.15 shows LQs for manufacturing employment and professional
employment for the Los Angeles urban area. A high-tech business that man-
ufactures and provides professional services can use maps such as these to
make decision on siting of its plants, R&D facilities, and office locations. For
a large corporation with locations in hundreds of cities worldwide, LQ spatial
analysis can lead to competitive advantage on a national or global scale.
NATIONAL SPATIAL DATA INFRASTRUCTURES
Governments have recognized the importance of spatial information by es-

tablishing National Spatial Data Infrastructures (NDSIs), sometimes referred
to as Spatial Data Infrastructures (SDIs) (Masser, 2005). Governments can
create national data infrastructures that apply to the nation as a whole. Early
adopters of NDSIs are Australia, Canada, Indonesia, Japan, Korea, Malaysia,
Netherlands, Portugal, Qatar, the UK, and the U.S. In the U.S, an NDSI vision
was set forward by the Office of Management and Budget in 2002 (Masser,
2005). The lead agency is the FGDC (Federal Geographic Data Committee) in
the U.S. Geological Survey the Department of Interior that supports the goals
of the U.S. NSDI of data sharing, use and dissemination nationally of spatial
data, and the interoperability of federal spatial systems (Masser, 2005). “The
NSDI is a physical, organizational, and virtual network designed to enable the
development and sharing of this nation’s digital geographic information re-
sources.” (FGDC, 2007). Since much of the spatial data of businesses comes
from government sources, the development and growth of NDSIs worldwide
is a significant trend.
GEOWEB
The user-friendly environment that supports less technically skilled web-

based users is called the GeoWeb (Dangermond, 2006; Lake, 2006). It has
been defined as “a large, widespread, distributed collaboration of geospatial
knowledge services that can be dynamically interrelated and orchestrated for
application support” (Dangermond, 2006). The GeoWeb involves the spatial
web services that are not just a group of isolated services acting separately,
but services collaborating with each other and exchanging information in
orchestrated steps. The GeoWeb is global and is not slowed by geographical or
jurisdictional boundaries (Dangermond, 2006; Lake, 2006). Good examples
of the GeoWeb have occurred in times of disaster. For instance, following
Hurricane Katrina in 2005, emergency spatial information was available as
a set of map web services. This made the information widely available to
a disrupted and scattered citizenry as well as to hard-pressed emergency
workers. Another example is the real estate industry, which has had strong
adoption of a variety of map web services, including such organizations as
book cases Sperry Van Ness, Prudential Preferred Realty, Trulia.com, and
Fanny Mae Foundation. These and others overlap in their APIs, mapping data
sources, and attribute data sources, and sometimes interact with each other.
The upshot is that the way of doing business in real estate has changed, for
instance many customers no longer need to take home tours. Figure 2.16
shows GeoWeb users in a corporation discussing a problem.
In the GeoWeb, map services are available for mapping, geocod-
ing, gazetteer (online atlases), viewing terrain in 3-D, image processing,
FUTURE TRENDS IN SPATIAL TECHNOLOGIES 65
Figure 2.16 GIS Specialists Working on a Problem
multimedia, sensor data, real-time monitoring, and cataloging of metadata

(Dangermond, 2006; Lake, 2006). The services are able to collaborate and to
exchange information. Behind this are agreements, standards, and procedures.
Along with movement in GIS and spatial applications towards the GeoWeb
environment, mainstream IT is becoming spatially-enabled. An important
dimension of this for enterprise systems will be seen in Chapter 4. ERP,
CRM, and other large-scale backbone IT systems were originally introduced
as non-spatial, but many now offer inherent spatial features or connectors to
spatial software. Even with the spatial features, these mainstay applications
are not centered on spatial, but rather on supporting the company’s underlying
business processes. The spatial features are available when needed to add
to the support. Likewise, for spatial decision support systems, the spatial
component is a complement to existing models. For instance in the LCI case
in Chapter 3, underwriters nationally utilize models to price policies that
include spatial features, but the system is principally IT-oriented. In fact, the
underwriters usually prefer not to look at maps, but benefit through spatial
analysis by having more accurate modeling outcomes. As spatial techniques
are only one one of multiple components in these models, these are IT systems
that are spatially-enabled rather than GIS (Barnes, 2001; Sonnen, 2006).
FUTURE TRENDS IN SPATIAL TECHNOLOGIES
Where is the evolution of spatial technologies heading? Among the future

trends are the following (Buyya and Venugopal, 2005; Dangermond, 2006;
Sonnen, 2006):
r GIS and spatial technologies will become commonplace in IT systems.

r Spatial Data Infrastructures (SDIs) will offer architectures to share geo-
graphic information nationally and internationally.
r The spatial centralized storage areas of enterprise systems in organiza-

tions will become larger and more efficient.
r More spatial and GIS software code will become openly available, lead-
ing to a more open source environment.
r Spatial standards will become more widely accepted, as well as vendor
compliance with standards.
r Satellite, sensor, and other spatial data gathered in real-time will be
pervasive and permeate real-time business systems.
r GIS-enabled and locationally-aware mobile technologies will become
prevalent.
r For the web infrastructure, greater collaboration, range, and intensity of
spatial web services will occur through movement towards the emerging
concepts of semantic web, spatial agents, and Grid computing. Grid
computing will be discussed in the end section of Chapter 5.
CHAPTER SUMMARY
The chapter has provided an overview of key concepts from Information

Systems that form one of the bases for GIS in business. These concepts were
related to GIS as a type of IS. The chapter then proceeded to provide more
background on spatial functioning including the point location, coordinate
systems, and forming map layers from points, lines, and polygons.
The linkages of GIS to other disciplines are important since a mixture of
concepts and theories are brought to bear in the business context. The case
study of Baystate Health concerns a successful health-care system that has
introduced GIS through fine leadership, development of professional staff,
carefully chosen spatial applications that are meaningful, and organizational
collaborations internally and externally.
The chapter examines the key topics for any GIS of data sources, entry of
spatial data, and spatial analysis. The latter is illustrated by a location quotient
example. It ends by considering Spatial Data Infrastructures, the concept of
GeoWeb, and the future trends in spatial technologies.
ACKNOWLEDGMENT
Part of the section on spatial analysis was published in an article of Commu-

nications of the Association for Information Systems (Pick, 2005). It is used
with the permission of the journal.
REFERENCES
Aoyama, Yuko, Samuel Ratick, and Guido Schwarz. 2006. “Organizational Dynam-
ics of the U.S. Logistics Industry: An Economic Geography Perspective.” The
Professional Geographer 58(3): 327–340.
REFERENCES 67
ArcNews. 2004. “On Demand, In Demand—ArcWeb Services.” Special Section.

ArcNews 7(4): 40–46.
ArcNews. 2005. “New Version of ArcWeb Services Allows Users to Customize
Services.” ArcNews 27(2): 1–3.
ArcUser. 2000. “Choosing the Best Projection.” ArcUser, 23.
Austin, Robert, and Stephen Bradley. 2005. The Broadband Explosion. Cambridge,
MA: The Harvard Business School Press.
Bolstad, Paul. 2005. GIS Fundamentals, 2nd Edition. White Bear Lake, MN: Elder
Press.
Brewer, Cynthia A. 2005. Designing Better Maps: A Guide for GIS Users. Redlands,
CA: ESRI Press.
Brewin, Bob. 2005. “No Silver Bullets.” Federal Computer Week, 19(13): 39,
May 2.
Chen, Y.H. 2004. “Getting Ready for RFID.” OR/MS Today (31)3, pp. 30–35.
Chilsholm, Patrick. 2005. “RFID.” Military Medical Technology. Online Edition.
http://www.military-medicial-technology.com/print article.cfm?DocID=431.
Clarke, K. 2003. Getting Started with Geographic Information Systems, 4th Edition.
Upper Saddle River, New Jersey: Prentice Hall.
Croft, John. 2005. “RFID: Be Not Afraid.” Overhaul and Maintenance, Washington,
D.C., April 1.
Dale, P.F., and R.A. McLaren. 2000. “ GIS in Land Administration.” In Longley, P.A.,
M.F. Goodchild, D.J. Maguire, and D.W. Rhind (eds.), Geographical Information
Systems, Two volumes, New York: John Wiley and Sons Inc., 859–875.
Dangermond, Jack. 2006. “GIS Enterprise Architecture: Unifying the Utility.” GITA
2006 Proceedings.
ESRI. 2005. “About ArcWeb Services.” Available at www.esri.com.
ESRI. 2006a. “ArcGIS.” Redlands, CA: ESRI Inc.
ESRI. 2006b. “ArcGIS 3D Analyst.” Redlands, CA: ESRI Inc.
ESRI. 2007. “MapObjects-Windows Edition.” Available at www.esri.com.
FGCD. 2006. Federal Geographic Data Committee Website. Reston, Virginia: FGCD.
Available at www.fgcd.gov.
Galuszka, Peter. 2005. “Lessons from Baghdad.” Chief Executive, Vol. 208, May,
p. 48.
Getis, A. 2000. “Spatial Statistics.” In Longley, P.A., M.F. Goodchild, D.J. Maguire,
and D.W. Rhind (eds.), Geographical Information Systems, Volume 1, New York:
John Wiley and Sons Inc., pp. 239–251.
Godin, L. 2001. GIS in Telecommunications. Redlands, CA: ESRI Press.
Gregory, I., K. Kemp, and R. Mostern. 2002. “Geographical Information and Histor-
ical Research: Integrating Quantitative and Qualitative Methodologies.” Humani-
ties and Computing.
Greene, R.P., and J.C. Stager. 2005. “Techniques and Methods of GIS in Business.”
In Pick, J.B. (ed.), Geographic Information Systems in Business, Hershey, PA:
Idea Group, pp. 36–55.
Greene, R.W. 2000. GIS in Public Policy. Redlands, CA: ESRI Press.
Guerrero, Ignacio. 2005. “Emerging Technologies in the Geospatial Industry.” In Pro-

ceedings of the Annual Conference of the Geospatial Information and Technology
Association (GITA), GITA: Aurora, CO.
Haag, Stephen, Maeve Cummings, and Amy Phillips. 2005. Management Information
Systems. New York: McGraw Hill.
Harder, Christian. 1997. Arcview GIS Means Business. Redlands, CA: ESRI Press.
Harder, Christian. 1999. Enterprise GIS for Energy Companies. Redlands, CA: ESRI
Press.
Heywood, Ian. 2006. An Introduction to Geographical Information Systems, 3rd
Edition. Upper Saddle River, NJ: Prentice Hall.
Hiremane, Radhakrishna. 2005. “From Morre’s Law to Intel Innovation—Prediction
from Reality.” Technology@Intel Magazine, April, 3–8.
Huxhold, William E. 1991. An Introduction to Urban Geographic Information Sys-
tems. New York: Oxford University Press.
Huxhold, William E., and A.G. Levinsohn. 1995. Managing Geographic Information
Systems Projects. New York: Oxford University Press.
Johnston, K., J.M. Ver Hoef, K. Krivoruchko, and N. Lucas. 2001. Using ArcGIS
Geostatistical Analyst. Redlands, CA: ESRI Press.
Karini, Hassan A., and Amin Hammad. 2004. Telegeoinformatics. Boca Raton, FL:
CRC Press.
Keenan, P. 2005. “Concepts and Theories of GIS in Business.” In Pick, J.B. (ed.),
Geographic Information Systems in Business, Hershey, PA: Idea Group, pp. 1–19.
Kolb, Karen. 2001. “Medical Center Improves Care, Prevention Programs with GIS.”
ArcUser 4(3): 24–25.
Laudon, Kenneth C., and Jane P. Laudon. 2007. Essentials of Business Information
Systems, 7th Edition. Upper Saddle River, NJ: Prentice Hall.
Lo, C.P., and K.W. Yeung. 2002. Concepts and Techniques In Geographic Information
Systems. Upper Saddle River, NJ: Prentice Hall.
Longley, P.A., M.F. Goodchild, D.J. Maguire, and D.W. Rhind. 2000a. “Introduc-
tion.” In Longley, P.A., M.F. Goodchild, D.J. Maguire, and D.W. Rhind (eds.),
Geographical Information Systems, Two volumes, New York: John Wiley and
Sons Inc., pp. 1–20.
Longley, P.A., M.F. Goodchild, D.J. Maguire, and D.W. Rhind (eds.). 2000b. Geo-
graphical Information Systems, Two volumes. New York: John Wiley and Sons
Inc.
Longley, P.A., and M. Batty (eds.). 2003. Advanced Spatial Analysis: The CASA
Book of GIS. Redlands, CA: ESRI Press.
Maguire, David J. 2006. “New Approaches for GIS-based Modeling and Analysis.”
GITA 2006 Proceedings.
Masser, Ian. 2005. GIS Worlds: Creating National Spatial Data Infrastructures.
Redlands, CA: ESRI Press.
Miller, Fred. 2007. GIS Tutorial for Marketing. Redlands, CA: ESRI Press.
Mitchell, A. 1999. The ESRI Guide to GIS Analysis Volume 1: Geographic Patterns
and Relationships. Redlands, CA: ESRI Press.
REFERENCES 69
Moore, John. 2005. “RFID’s Positive Identification.” Federal Computer Week 19(11):
53, April 18.
OECD. 2007. “OECD Broadband Statistics to December 2006.” Directorate for Sci-
ence, Technology, and Industry. Paris: Organisation for Economic Co-operation
and Development. Available at www.oecd.org/sti/ict/broadband on June 15,
2007.
Oliva, Lawrence M. 2004. IT Security: Advice from Experts. Hershey, PA: CyberTech
Publishing.
O’Sullivan, David, and David Unwin. 2003. Geographic Information Analysis. Hobo-
ken, NJ: John Wiley and Sons.
Pick, James B. 2004. “Geographic Information Systems: A Tutorial and Introduction”
Communications of the Association for Information Systems, 14: 307–331.
Ray, J. 2005. “Spatial Data Repositories: Design, Implementation, and Management
Issues.” In Pick, J.B. (ed.), Geographic Information Systems in Business, Hershey,
PA: Idea Group, pp. 80–112.
Reeve, D.E., and J.R. Petch. 2002. GIS, Organisations and People: A Socio-Technical
Approach. London: Taylor and Francis.
RFID Journal. 2003. “Military Edict: Use RFID by 2005.” RFID Journal, October 3.
Available 5/12/05 at www.rfidjournal.com.
RFID Journal. 2003. “U.S. Military to Issue RFID Mandate.” RFID Journal,
September 15. Available 5/12/05 at www.rfidjournal.com.
Satzinger, John W., Robert B. Jackson, and Stephen D. Burd. 2004. Systems Analysis
and Design, 3rd Edition. Boston: Thompson Course Technology.
Smith, Gary, and Joshua Friedman. 2004. “3D GIS: A Technology Whose Time Has
Come.” Earth Observation Magazine, November.
Sonnen, David, and Henry D. Morris. 2005. “Oracle 10g: Spatial Capabilities for
Enterprise Solutions.” IDC White Paper. Framingham, MA: International Data
Corporation.
Sugumaran, Ramanathan, Shiram Ilavajhala, and Vijayan Suguraman. 2007.
“Development of a Web-Based Intelligent Spatial Decision Support System
(WEBISDSS): A Case Study with Snow Removal Operations.” In Hilton, Brian
(ed.), Emerging Spatial Information Systems and Applications, Hershey, PA: Idea
Group Publishing, pp. 184–202.
Sun Microsystems. 2005. “RFID Keeps U.S. Troops Well Stocked.” Available
5/12/2005 at http://www.sun.com/solutions/documents/articles/go rfid dod.xml.
Tao, Vincent. 2003. “The Smart Sensor Web.” GeoWorld, 16(9): 28–32.
UCGIS. 2007. University Consortium for Geographic Information Science. Website.
Available at www.ucgis.org.
U.S. Geological Survey. 2002. Geographic Information Systems. Washington, D.C.:
U.S. Geological Survey.
CHAPTER 3
SUPPORTING BUSINESS DECISIONS
INTRODUCTION
Spatial technologies are growing rapidly and becoming more web-dependent

and competitive. Information systems today support the enterprise almost
from top to bottom as an essential element (Gray, 2006). GIS rests on the
IS foundation and as well on geospatial concepts and methods. IS and GIS
are linked together through common elements such as database, networking,
systems development, and IS organizational and management principles. On
the other hand they differ in GIS’s recognition of location, mapping, and
IS’s greater scope. GIS and IS have in common their support for business
decisions.
GIS and spatial technologies have a crucial role in supporting people in
businesses to make decisions. This resembles IS’s longer established role in
business decision-making. The IS field has well-known systems for decision-
support including decision support systems (DSS) and business intelligence
(BI) (Marakas, 2002; Power, 2002; Gray, 2006). For GIS, there are the rela-
tively new concepts of spatial decision support systems (SDSS) and spatial
business intelligence. An example of these concepts is models and data that
help bank managers overseeing a rapidly expanding branch network to make
better decisions on locating some branches, while closing, merging, or relo-
cating others. These decisions are based on spatial displays and analysis of
the customer base, the competition, and branch locations.
This chapter concerns GIS and spatial technologies applied to business
decision-making. It starts with the concepts of decisions, decision support
70
GIS TO SUPPORT DECISION-MAKING IN ORGANIZATIONS 71
systems, and spatial decision support. A case study of spatial decision support
in the U.S. division of a global bank is given. The chapter then looks at spatial
decision support systems and spatial business intelligence (spatial BI), seeking
to explain their concepts, models, and applications. It finishes with a case on
spatial decision-making in a large U.S. insurance firm.
GIS TO SUPPORT DECISION-MAKING IN ORGANIZATIONS
Businesses depend on continual decisions. Executives guide the enterprise in

deciding on long-term strategies, while middle managers decide on tactical
initiatives to achieve middle range project goals. Knowledge workers make
decisions to analyze business problems, conduct research, develop products,
and generate creative decisions to put the enterprise in the forefront. Op-
erations supervisors and managers focus on proximate decisions to operate
particular business processes or sub-processes. For any organization, deci-
sions constantly appear and need to be acted upon (Marakas, 2002).
In IS, decision support systems (DSSs) have been available since the 1980s
and continue to evolve into new versions that incorporate the web, intelligent
agents, and mobile technologies. A DSS is a software package that has a
database, model base, knowledge engine, and user interface that are pro-
grammed to support management decisions (Marakas, 2002; Gray, 2006).
The concept of DSS is necessary to review first, before moving to the major
topic of SDSS.
A DSS is best suited for semi-structured or unstructured decisions (Gorry
and Scott Morton, 1989; Marakas, 2002). A structured decision implies that
the steps to reach it are logical and can be automatically calculated. For
instance, the accounting calculations to produce a company balance sheet are
logical and automatic. A person does not have to make the decisions, and the
steps can be delegated to a computer.
An unstructured decision occurs in situations where the objectives of a
decision conflict with other decisions, the alternatives for a decision are not
clear, and/or the impacts of a decision are not known (Power, 2002; Marakas,
2002). An unstructured decision cannot be exactly calculated because its
mechanisms are not known or it is too complex to model. For example,
the decision on the strategy of a political campaign cannot be automated
into fixed steps. Between structured and unstructured decisions are semi-
structured ones, which can be partially calculated but not entirely. Some parts
of the decision can be automated, so human intuition must comes into play.
An example of semi-structured is the decision on how much to spend to
construct a new warehouse. A DSS addresses semi-structured or unstructured
decisions by sizing up the problem, selecting the appropriate model from a
group of models referred to as the model base, and running the model which
accesses data from the database (Turban and Aronson, 2001; Marakas, 2002).
72 SUPPORTING BUSINESS DECISIONS
External Internet,
Other computer- Intranets,
and based systems
internal Extranets
data
Model Knowledge-based
Data Management
Management Models
User Interface
Manager or
Knowledge base
other user
Figure 3.1 Diagram of a DSS. Source: Turban and Aronson, 2001
A DSS depends on a database management system that manages the stor-

age, retrieval, and security for the data. As seen in the schematic diagram in
Figure 3.1, the model management system makes available one or more mod-
els from a stored set of them. They can be quite varied (Turban, 1995; Marakas,
2002). Some important ones are optimization, simulation, and what-if (see
Table 3.1). Depending on the decision being made, the model base determines
which model to invoke, gathers the relevant data, and accesses a knowledge-
based model if appropriate. A knowledge-based model includes a knowledge
engine that applies logical reasoning based on rules, tasks sometimes called
expert systems. It stores and utilizes logical rules, constraints, and heuristic
principles (Marakas, 2002). The rules are extracted from human experts and
stored in the knowledge base. The last major component, the user interface
(sometimes called graphical user interface or GUI) provides the interactive
interface for the user to communicate back and forth with the DSS. Through
the user interface the user can interact with the data management, model
management, and knowledge-based modules. It is crucial because the typical
DSS user, a manager or executive, has limited time to communicate and is
usually not a technical expert. As an example of a DSS use, a telecommuni-
cations manager who needs to estimate customer demand interacts with the
user interface to call up an exponential-smoothing forecasting model from
the model base which in turn invokes the data management system to draw
data from a customer database. The basic components of a DSS are relevant
to the chapter since they carry over to SDSS.
A decision support system assists in decisions at varied organizational
levels, from the top executive down to supervisors of factory processes and
clerical groups. As seen in Figure 3.2, executive management makes strategic
TABLE 3.1 Types of Decision Support Systems Models
Category Process and Objective Representation Techniques
optimization (few find best solution from few decision tables, decision trees
alternatives) alternatives
optimization (many find best solution from many mathematical programming,
alternatives) alternatives by iterative network models
improvement
analytical solution to find best solution in one step by some inventory models
optimization formula
simulation mimic reality to find a “good” wide variety of models in
solution (perhaps best) nearly all fields
heuristics use practical rules to find “good” expert systems
solution
predictive models forecast the future based on forecasting models, Markov
(forecasting) assumptions models
“what if” models vary assumptions to assess financial modeling
effects on outputs/sensitivity
analysis
Source: Turban, 1995
Figure 3.2 Organizational Hierarchy for SDSS and Spatial BI. Source: Manglik,
2006
decisions that are long-term and have enterprise-wide impacts. The data pro-
vided to top executives tend to be aggregated and integrated across the com-
pany. For instance, a CEO accesses the historical time series of data on all the
firm’s customer service requests in North America versus Europe, in order to
gauge company-wide responsiveness. The executive-level data are often ex-
ternal, since the executive needs to environmentally scan and assess facts and
forces outside the firm. Executives, who often have limited time and reduced
hands-on experience, need a friendly user interface. Such a sub-category of
DSS focusing on top executives is referred to an executive information system.
Middle managers, with responsibility for divisions, departments, or re-
gions, make periodic decisions that relate mostly to tactical initiatives af-
fecting their organizational unit. They need partially aggregated data, and
they review reports that are often periodic. In addition to deciding on busi-
ness processes in their unit, their swath of decisions also cuts across other
units, so they sometimes require integrated information. Knowledge workers
make decisions in business planning, forecasting, R&D, and creative areas,
with little direct impact on operations. They base their decisions on a range
from specialized to aggregate information, which may be internal or exter-
nal depending on the analytic problem. Finally, operational supervisors and
managers need support for real-time decisions involving a business process
or sub-process in their unit. For instance, a utility field supervisor benefits by
decision support on how to pattern the next week’s maintenance tasks on a
segment of pipeline. The data are specialized, recent, and keyed to the unit’s
immediate needs.
Other types of systems besides DSS impact all of these levels, so the
role of DSS needs to be pinpointed more. By cross-classifying the level
of management activity with the structuredness of the decision (Table 3.2),
DSS is seen to be especially strong in supporting semi-structured decisions
at the strategic, management levels, and knowledge levels (Gorry and Scott
Morton, 1989; Marakas, 2002). DSSs may fail when confronting unstructured
decisions, because their model base cannot provide an appropriate model
leading to decisions based on experience and “gut instinct.”
In summary, DSSs can support a variety of managers in making semi-
structured and/or somewhat unstructured decisions. A DSS can support an
individual or group in decision-making. The latter type of system is known as a
GDSS—group decision support system (Marakas, 2002). Unlike transactions
processing and management information systems, the DSS has the capability
to respond to surprises and unprecedented situations, and can support simul-
taneous decision-making at different organizational levels and geographies in
the company.
Business Intelligence (BI) is closely related to DSS, but is broader and
draws from more diverse sources. A BI is a combination of data gathering,
data storage, and knowledge management with analysis and modeling tools
that provides complex enterprise-wide corporate and competitive information
to decision makers (Gray, 2006). The BI gathers more far-reaching and diverse
information and tends to be focused on more complex situations than a DSS.
TABLE 3.2 Classification of a DSS
Decision Type Structured Semi-Structured Unstructured
Management
Activity
Operational Control Inventory control Securities trading Decision on cover
photo for monthly
magazine
Knowledge Consulting report Research on market R&D on emerging
Management on standard changes for fast technologies
production foods in a region
process
Management Load balancing of Establishing marketing Hiring of management
Control production lines budgets for new personnel
products
Strategic Planning Physical plant Analysis of acquisition Determining what
location of capital assets R&D projects to
undertake
Support Needed Transaction DSS Executive Information
Processing System (EIS),
Systems, MIS human reasoning
and intuition
Source: Reprinted and adapted from A Framework for Information Systems by A. Gorry and M. Scott-
Morton, MIT Sloan Management Review, Vol. 13, 1991, pp. 56–79, by permission of the publisher.
Copyright © 1971 by Massachusetts Institute of Technology. All right reserved.
Like DSS, BI has components for data management, model management,

knowledge management, and a user interface. Its knowledge management
function is broader and more powerful than for a DSS. In the book, we regard
DSS and BI as close relatives that share many features and constitute the
ends along a continuum of decision systems. A further distinction separates
the vendor products in DSS, BI, SDSS, and spatial BI. Vendor products fall
along the DSS-to-BI range described, but recently vendors have tended to
prefer their products to be called “BI.” In the book, when vendor products are
discussed, the vendor’s own terminology (DSS or BI) is respected.
A spatial decision support system (SDSS) is a DSS that provides locational
decision support where there is a geographic or spatial component in the
decision-making (Keenan, 2005; Jarupathirun and Zahedi, 2005). Likewise a
spatial business intelligence system (spatial BI) is a BI that provides locational
decision support where there is a spatial element in the decision-making. The
generic SDSS seen in Figure 3.3 falls in between GIS and DSS (Huerta,
Navarrete, and Ryan, 2005). The same added components of spatial analysis
and spatial boundary layers apply to Spatial BI, which falls between GIS and
BI.
An SDSS combines GIS and DSS features. It may be formalized into a
single system or product, or reside in a combination of two or more loosely
GIS SDSS DSS

Systems that Systems that Systems that
collect, store, support decision- support the
analyze, and making when decision making
display spatia spatial data are process
data involved
Figure 3.3 GIS, DSS, and SDSS. Source: Huerta, Navarrete, and Ryan, 2005.
Copyright 2005, IGI Global, www.igi-pub.com, reprinted by permission of the pub-
lisher.
coupled products. In the latter case, a DSS vendor may offer a product that
interfaces with a GIS package. Later in the chapter, the BI product WebFocus
and loosely coupled GIS product ArcGIS are described.
A conceptual model of a typical SDSS (Figure 3.4) enlarges the DSS model
from Figure 3.1 by adding spatial analysis and spatial data components. The
spatial analysis model performs the specialized spatial functions discussed
in Chapter 2 such as buffer, overlay, proximity estimates, distance measure-
ment, longitudinal change analysis, spatial statistics, and spatial econometrics.
These techniques perform computations based on locations and geographic
surfaces. The spatial data are the points, lines, and polygons arranged in
boundary layers from Chapter 2.
Internet,
External Other computer-
Intranets,
and based systems
Extranets
internal
data
Data Model Knowledge-based

Management Management Models
Spatial Analysis User Interface
Manager or
Knowledge base
other user
Spatial Boundary
Layers
Figure 3.4 Diagram of an SDSS.
SDSS may be viewed along the continuum of GIS to SDSS to DSS (see
Figure 3.3). Thus some SDSSs may resemble a GIS more than a DSS, while
others are closer to DSS but with spatial features. In Chapters 1 and 2, it was
emphasized that GIS is defined as having analysis capabilities. This version of
an SDSS (in the left-middle in Figure 3.3) has an enlarged analysis component
of GIS that serves for decision support.
Historically, a line of GIS and products has grown and developed with-
out recognizing formal DSS concepts or DSS products. This line has led
to today’s GIS decision-modeling products. For example, ArcGIS’s Model
Builder feature allows the user to build a model interactively and visually by
designing the model layout of attributes and their relationships, and testing if
the model components work together. It produces spatial analysis and maps,
based on the model’s logical flows (ESRI, 2006).
The spatial decision systems derived from the GIS side are important and
dominant for certain companies and industries. This chapter’s large com-
mercial bank case illustrates the GIS approach. Another example of the GIS
approach to decision support occurs in the utility industry discussed in Chap-
ter 9. The GIS-derived decision support assists utility engineers in planning
and design of networks and facilities. Also, GIS assists decision-making in
site selection, meter readings, fleet management, outage management, demo-
graphic analysis, and logistics. In health care, Kaiser Permanente, a national
hospital chain headquartered in California, follows GIS for decision support,
basing decisions on spatial analysis and modeling to classify in-service ar-
eas, profile disease patterns of regions, analyze optimal distribution of health
care resources such as physicians and MRI equipment, and spatially assess
facility siting. What decision support approach should a company practically
adopt—(1) that derived from DSS and BI products or (2) that from GIS prod-
ucts? It depends on such factors as staff training and education, how inherently
geographical the decision problems are, prior adoption of GIS or DSS/BI soft-
ware that can be built upon, the organizational reach of the system (depart-
mental versus enterprise), and the cost. It is recommended that a firm trying
to decide on the appropriate approach perform a cost-benefit analysis of the
alternatives, a topic in Chapter 7. Large organizations and technically inclined
ones might adopt both approaches for different business units and projects.
The history of spatial decision support has unfolded over four decades.
DSS was conceived with pioneering work in the late 1960s and 1970s, much
conducted at MIT, where Gorry and Scott Morton (1971) originally identified
non-structuredness as crucial and outlined its framework given in Table 3.2.
The GIS approach began in the mainframe era of the late 1960s and evolved
through client-server into modern forms. The parallel evolution of GIS and
DSS systems (Jarupathirun and Zahedi, 2005) is shown in Table 3.3. An early
SDSS paper introduced the GADS (Geodata Analysis and Display System),
an IBM training system for routing decisions that included maps. Starting
in the mid 1970s (Nagy and Wagle, 1979), decision support based on GIS
evolved and became recognized in the GIS and GI Science fields. In the
TABLE 3.3 Parallel Stages in Evolution of GIS and Decision Support Systems (DSS)
time
Mobile GIS Mobile DSS
Intelligent GIS Intelligent DSS

(e.g. personalizing GIS based (e.g. intelligent agents for
on users' abilities and needs) helping in decision making
Web-based GIS Web-based DSS

(e.g. using GIS for helping web- (e.g. using DSS for helping
users find locations, directions web-users in their financial
or distances) decisions or choice of products)
Advanced GIS Decision Support Systems

(including spatial databases, (including model bases, databases,
spatial statistics models, and interactive and user-friendly interfaces)
interactive interfaces)
GIS for Map Processing Transaction Processing Systems

(e.g. off-line map generation) (e.g. off-line report generation)
Source: Jarupathirun and Zahedi, 2005. Copyright 2005. IGI Global. www.igi-pub.com, reprinted by
permission of the publisher.
1980s and 1990s, advanced visualization was incorporated into GIS. SDSS
was included as a research and curricular component in the first U.S. Center
for Geographic Information and Analysis, founded at University of California
Santa Barbara in 1990.
By the mid 1990s, web-based DSS and GIS had become available and
improved rapidly over the last decade. Subsequently, SDSS and spatial BI
have become internet-enabled and often implemented on the enterprise-wide
internet bus (Figure 1–11). In recent years, these systems have become avail-
able in mobile devices and in conjunction with coupled spatial technologies.
Expert systems, which emphasize rule-based logic, appeared originally in
the 1960s from research on artificial intelligence, and are now incorporated
into DSS, SDSS, BI, and spatial BI as part of the knowledge management
component (Marakas, 2002). Intelligent agents are software packets, often
having artificial intelligence features, that act on behalf of the user to perform
repetitive computer-related tasks (Marakas, 2002; Wikipedia, 2006). Some
SDSS and spatial BI systems have incorporated intelligent agents (ESRI,
2006; Sugumaran et al., 2003, 2004, 2005). Known in simple terms as “wiz-
ards,” intelligent agents can relieve the decision user of the need for complex
coordination and command sequencing of data, models, and software.
However, for expert systems and intelligent agents, there is a tradeoff.
Although they relieve the user of expert knowledge and complex coordination,
CASE STUDY OF SPATIAL DECISION-MAKING IN A LARGE PERSONAL/CORPORATE BANK 79
they also shield him/her from knowledge and control of the detailed decision
processes (Jarupathirun and Zahedi, 2005). Today there are a variety of spatial
decision-making approaches, ranging from commercial to in-house, from
standalone to coupled packages, and from the GIS modeling approach to
SDSS and spatial BI. The future evolution of spatial decision-making support
is uncertain, but is likely to move in the direction of the web/internet, mobile,
and coupled spatial technologies.
CASE STUDY OF SPATIAL DECISION-MAKING IN A LARGE

PERSONAL/CORPORATE BANK
This case concerns spatial decision-making at the U.S. division of Large

Personal/Corporate Bank (LPCB), a global company headquartered in
Europe. The U.S. division has grown rapidly in the Northeast and some
Sunbelt locations. With an emphasis on personal and corporate customers, its
expansion is driven by its branch network, which numbers in the mid hun-
dreds. The support for spatial decision-making at LPCB is focused on helping
middle managers and knowledge specialists understand the trade areas, cus-
tomer profiles, and marketing potential in the areas where its branches are
already located and new potential branch zones.
LPCB’s GIS team consists of a manager and two specialists. It resides
organizationally in the personal banking division, where it is part of the office
of branch location strategy. The firm’s IT group provides loose support to the
GIS team’s hardware and software needs. The GIS team’s software from ESRI
and MapInfo resides on a small client-server network that is not web-based.
The decision-support approach taken is to utilize the analysis and modeling
tools of commercial GIS software packages, but in certain projects to run a
commercial location planning software package, that loosely interacts with
the GIS. The GIS team in the U.S. has limited contact with GIS teams in other
countries, which vary in size and sophistication.
The bank’s GIS decision-support applications are trade area analysis for
branches and ATMs, market demographics and competitor analysis, branch-
related campaigns, location allocation, forecast of market potential for LPCB
securities, and mapping to support executives in their strategies.
(1) Trade Area Analysis

The bank collects extensive customer data down to the Census block-group
level, which are utilized to gain understanding of what branch a customer is
assigned to, at which bank he/she opens accounts, the volume the customer is
transacting, and where transactions originate. A block group is a U.S. Census
small-area geographic designation that contains about 1,500 residents. It is
common that a customer will open an account at one location but make all
his/her transactions at others. The analysis takes this into account and gives
a picture of the customer market around branches, so decisions can be made

on retaining, in-filling, re-aligning, and closing branches and ATMs.
(2) Market Demographics, Competitor Analysis,

and Customer Satisfaction
Detailed analysis can be done of LPCB’s market share and the location of
competitors’ versus LPCB’s branch locations. It pinpoints where there are
high-value customers who are not served. Customers’ ratings of customer
service are analyzed with attention to spatial clustering. For instance, a very
busy branch may have low customer service scores because of the waits and
delays. This analysis can point to remedies such as locating new branches or
adding to the existing branch’s capacity.
(3) Branch-Related Campaigns

LPCB frequently conducts campaigns to target the locations of new prospects
and markets. Spatial analysis helps to focus and speed the campaigns. A GIS
team member stated, “If you hit the right people in the right areas, you get the
best return on the dollar.” GIS is also used to assess the strength of a branch
manager relative to a particular campaign in his/her area.
(4) LPCB Securities

Securities and insurance sales representatives at the bank utilize geographic
planning tools to gain insight on the market potential for their products.
The bank’s internal geographic allocation of securities/insurance people and
resources can be assessed versus the nearby market potential.
(5) Mapping Support for Executives on Strategy

Senior management at LPCB/USA are supported by spatial analysis regarding
the bank network, the market potential of products and services, and the
competition, in order to gain a general understanding of the bank’s situation
and its opportunities.
The SDSS approach for LPCB/USA consists of GIS with enhanced analy-
sis, rather than SDSS or spatial BI software. The GIS approach is broad-based.
It utilizes customer, competitor, and demographic data, financial traffic pat-
terns in branches, and business patterns in shopping centers, It ties all the
items together to support and resolve decision-making. GIS is regarded as
providing an objective and broad-based view that serves as a “third party
voice” among many contributors to the decision process. Decision support
focuses on marketing managers, regional managers, and some specialists.
The decision support is especially important for allocations and re-
alignments of branches. Here bank managers have become increasingly
SPATIAL DECISION SUPPORT SYSTEMS 81
dependent on the objectivity of spatial analysis. GIS has become a formal

part of business procedures for decisions on whether or not to open new
branches. It’s also crucial to ATM locational decisions. The GIS team’s in-
puts go into every branch decision, but the team is not the “power broker,”
rather it is up to senior management to pick and choose what information they
want and ultimately the decisions. At the same time, a secondary role of the
GIS team is to provide senior management with an overall perspective on the
branch network.
Although senior management has not yet recognized GIS in the bank’s
strategic plan, GIS has moved up in importance indirectly through its impact
on critical strategic factors of branch expansion.
SPATIAL DECISION SUPPORT SYSTEMS
This section further explains SDSS, examines an academic conceptual model

of SDSS, considers the visualization aspect, looks at its prevalence and the
barriers to SDSS use, and considers an example of SDSS for typhoon insur-
ance.
In the general model of SDSS (Figure 3.4), it is helpful to add human
dimensions to the components of spatial analysis, spatial data, and the user
interface, going beyond the software to consider the spatial tasks people un-
dertake, user abilities to succeed with the tasks, user goals, and performance.
Companies and their people often have less familiarity with spatial appli-
cations, so the user’s motivation and his/her fit with the tasks is critical for
success. In the LPCB case, for example, there was positive motivation by the
GIS team and a good fit between the spatial analysis methods and the branch
marketing tasks that needed to be performed.
The conceptual model by Jarupathirun and Zahedi (2005) in Figure 3.5
is useful in enlarging the perspective to include the user, motivation, and
performance/utilization. The GIS Technology at the upper left represents
the map functions, modeling tools, and data management that appeared on
the left side of Figure 3.4. The tasks that utilize the technology may range
from simple to complex based on the user’s spatial abilities. The abilities
may be an individual’s or a team’s, as in the LPCB case. In Figure 3.5, user
motivation starts with self-perception of the tasks (“intrinsic incentive”) which
leads to goal commitment and establishment of a goal level. The amount
of commitment and aspiration in turn influences SDSS performance and
utilization. All this comes together in “task-technology fit,” which measures
whether the SDSS performance and utilization are commensurate to the size
and complexity of the decision tasks being confronted. Success depends on a
good balance of task and technology (Jarupathirun and Zahedi, 2005).
The user’s incentive, motivation, and goal setting are often overlooked
factors in considering spatial decision support. In the LPCB case, the team’s
positive incentives and motivation, and clear goal setting helped it to succeed.
GIS technology
- Simple map functions
Task-Technology
- Advanced tools
Fit
Geographical Tasks
- Simple
- Complex
Performance System Utilization
Spatial Abilities
Intrinsic Incentive
- Perceived required effort
- Perceived accuracy of Goal Commitment Goal Level
outcome
Figure 3.5 Conceptual Model of SDSS Task-Technology Fit. Source: Jarupathirun

and Zahedi, 2005 Copyright 2005, IGI Global, www.igi-pub.com, reprinted by per-
mission of the publisher
This area calls for an expanded research effort to understand how it occurs and
what its amount of impact is. Business management needs to gain awareness
of it and work at motivating/incentivizing GIS workers and teams.
An important model component is the user’s spatial abilities. Does he/she
have the natural ability to visualize maps and spatial relationships? Has he/she
received sufficient training and education in spatial software and technolo-
gies? The total amount of user ability needs to be assessed and steps taken
to bring the user or team up to a satisfactory amount. Studies have been
done of people’s inherent spatial ability to visualize and how that impacts
decision-making, Overall, the task-technology fit model (Jarupathirun and
Zahedi, 2005) is useful in broadening the dimensions of SDSS to extend to
the human spatial ability, motivation, and performance.
SDSS and Visualization

Visualization refers to the visual display of spatially arranged elements. The
user can better understand spatial relationships and assimilate new concepts
through visualization (Gonzales, 2004; Gershon and Page, 2001). People
think and communicate visually (DiBiase, 1990). This is evident today in the
prevalence of video, multimedia, and digital imagery. Visual understanding
helps in decision-making. Visualization can be emphasized and the outcomes
of spatial analysis made more meaningful through display of buffers, contours,
TABLE 3.4 Classification of Spatial and Non-Spatial Information Systems
Visualization Analysis
Spatial Attribute Spatial Attribute
Non-Spatial Information Systems

Transaction Processing Systems X
Management Information Systems X X*
Decision Support Systems X X
Spatial Information Systems
Consumer-Based Spatial Web Sites X X X* X*
Geographic Information Systems X X X X
Spatial Decision Support System X X X X
Note: * the analysis capabilities are limited when compared to decision support systems
Source: Copyright 2005, IGI Global, www.igi-pub.com, reprinted by permission of the publisher
graphics, and overlays; in 2D or 3D; and including satellite imaging, digital

photographs, or animation.
Spatial visualization is not a standard aspect of transaction processing
systems, management information systems, or decision support systems.
Table 3.4, modified from Jarupathirun (2005) illustrates in the left-hand col-
umn that spatial visualization is typical of consumer-based spatial web sites,
GISs, and SDSSs. The two right-hand columns indicate that spatial analysis is
not typical of transaction processing systems, MIS, and DSS, but is available
in the three categories of spatial systems, although less so for spatial websites.
These classifications underscore that spatial visualization is one of the defin-
ing characteristics of an SDSS and along with spatial analysis distinguishes
SDSS from DSS.
Experimental studies for over a decade have shed light on how visualiza-
tion helps for certain tasks depending on the profiles of users (Smelcer and
Carmel, 1997; Swink and Speier, 2000; Mennecke et al., 2000; Gershon and
Page, 2001; Kraidy, 2002). An early experiment (Smelcer and Carmel, 1997)
indicated that users performed faster problem solving with maps versus tables.
For both maps and tables, the time to solve problems grew as the difficulty
of the task increased. However, when users were tested on map-versus-table
understanding of proximity and adjacency geographic relationships, maps
encouraged short problem-solving time, whereas tables led to dramatically
increased time (Smelcer and Carmel, 1997). For a geographic tasks of contain-
ment, maps and tables performed the same. The practical implication of this
study is that for low-difficulty tasks, users performed the same for maps and
tables, but for greater task complexity, maps make a difference. This suggests
the advantages of SDSSs over DSSs for situations of high complexity.
In another experimental study (Swink and Speier, 1999), subjects’ perfor-
mance of working with a GIS was evaluated for different degrees of task prob-
lem size, data aggregation, and data dispersion. The subjects’ spatial-cognitive
skills were assessed by a “spatial orientation-card rotations test” (Swink and

Speier, 1999). They showed that spatial orientation skill leads to superior
quality of decisions. The user’s ability to understand data dispersion led to
superior decision quality. At the same time, the user performance with GIS
declined as the problem size grew. Although many cognitive factors were
left out, the study is useful in supporting that different users have different
inherent levels of spatial abilities.
An elaborate study (Mennecke et al., 2000) showed that experts are more
accurate than novices in utilizing GIS to perform geographic tasks. Experi-
ments on 240 subjects, students and professional planners, measured cognitive
fit, map interpretation, and reading. The students and professionals showed
more efficiency with SDSS compared to maps, when they worked on complex
problems, However, the professionals were more accurate yet less efficient
than students. In all, the investigation confirmed that there were decision gains
from SDSS compared to paper maps, particularly for more complex projects.
These and other studies of spatial visualization and cognition reveal gains
from GIS and spatial analysis versus tables and paper maps. They also show
differences in abilities for specific profiles of users and types of tasks. They
inform the topic of SDSS and spatial decision-making by highlighting the
user’s cognition and abilities as among the key success factors.
Prevalence of SDSS
Given the potential benefits of SDSS, it is crucial to ask how prevalent it is.
In lieu of any published data, examining the book’s research case studies can
provide an indication, even though they do not randomly represent a full pop-
ulation of firms with spatial applications. For this discussion, the firms are di-
vided into three size categories, and the results compared by size and industry.
For the ten large firms, only three had formal SDSSs. None of the three
used a standard SDSS or spatial BI package, but rather they utilized the mod-
eling and analysis features of standard GIS software for decision support.
Global Integrated Oil used standard GIS for decisions on facility location,
drill siting, and leasing locations. URS applied customized ESRI ArcGIS
product for decisions on managing groundwater and hazardous substances,
while Rand McNally applied Plan-o-grams and GIS software for decisions on
routing of sales reps, schedules, and product choices. The large Property and
Casualty Insurance Firm had an SDSS package that was written in-house and
based on simple knowledge rules. The SDSSs only interacted minimally with
other major business systems such as ERP, Customer Relationship Manage-
ment, and Supply Chain. In large firms, the users of spatial decision support
were mostly by managers and specialists, with reduced use by executives or
operational people.
For the five medium firms, there was moderate use but not of formal
software. None of these firms had formal SDSSs and only one of them sup-
ported decisions with enhanced GIS. However, all of the medium-sized firms
indicated that spatial data and programs support decision-making in areas
including inventory management, sales, advertising, newspaper circulation,

and hospital services. The users were similar to those of large firms—mostly
managers, and lesser use by executives and clerical staff.
For small firms, there is moderate decision support, but no formal SDSSs,
although one company provided SDSSs for clients in the form of GIS software
enhanced by macros and add-ons. Users at the client organizations were
managers, administrators, specialists, operations, and clerical people.
In sum, a sample of case companies had widespread use of spatial software
and programs for making decisions, but the use was largely not based on
formalized decision models. Where there was formalized spatial decision
support, it tended to be from GIS with strong decision modeling, rather than
from SDSS or spatial BI. Users tended to be mostly managers and specialists,
but there was some use by executives at large firms and clerical/operations
people at small ones.
Given the reduced formalized use of spatial decision support and BI indi-
cated, it is important to focus on costs and benefits, and ask what precludes
the formalized approaches from being adopted.
In the standard cost categories, formalized decision support accentuates he
costs of training, licensing, and consulting. More training is needed because
BI software skills are drawn from business and information systems fields,
more than geography and GIS. A concern was expressed about the “declining
depth of both conceptual knowledge and practical skills in computer sci-
ence/information technology among those individuals entering the geospatial
workforce” (Marble, 2006). SDSS and spatial BI are more easily adopted by
users trained in decision support systems and business intelligence, coming
from information systems and quantitative business degree programs. Use of
strong decision modeling components in a GIS also requires some of this
business understanding.
Licensing also potentially increases cost. SDSS and spatial BI products
are expensive—often $10,000 or much more per single-user license that
commonly is on top of GIS licensing cost. Start-up issues for spatial decision
support can be mitigated by skilled consultants, but they are expensive. On the
benefit side, it may be more difficult to justify the expense of formal spatial
decision support, because of its lack of a track record and less tangible pluses.
Since the SDSS users tend to be managers making tactical and sometimes
strategic decisions, it is harder to quantify the benefits.
A further reason formalized SDSS and spatial BI has not yet become
prevalent relates to their early growth stage. Richard Nolan (1973; 1979)
proposed six stages of information systems growth, as measured by growth
of computing budget. They are as follows: (1) initiation (acquisition of the
technology), (2) contagion (intensity in developing use and growing user
base), (3) control (reaction to contagion by imposition of accounting controls),
(4) integration (orientation towards user services), (5) data administration
(tight control of organizational computing resources, but some “slack” present
to develop systems with added value), and (6) maturity (final state of costs
under control and rising benefits). The stage theory is still relevant today,
E
Budget for Computing
D
A
Initiation Contagion Control Integration Data Maturity
Administration
STAGES
Figure 3.6 Nolan’s Stage Theory—1979 Version. Source: Reprinted by permission
of Harvard Business Review. This article was originally published under the title,
“Managing the Crisis in Data Processing” by Richard L. Nolan in March/April 1979.
Copyright © 1979 by the Harvard Business School Publishing Corp.
and is here applied to SDSS and spatial BI (see Figure 3.6 adapted from
Nolan, 1979). In the figure, the case-study responses would assign the SDSS
and spatial BI to late Stage 1 awaiting Stage 2. As a new form of product
having only been introduced five to ten years ago, it has not yet caught on.
By contrast, some newer spatial products have already reached contagion
(Stage 2), such Google Earth and related “mash-ups” covered in Chapter 5,
and navigation systems in cars.
Although the Nolan model was criticized as being only classificatory with-
out dynamic transition (King and Kraemer, 1984), it is useful in the context
of this book to gauging the maturity of spatial technologies. Stage theories
derived from Nolan’s work are in prominent use today including for strategic
information systems (Galliers and Sutherland, 2003) and enterprise systems
(Ross et al., 2006). This book is mainly intent on classifying maturity stages,
and does not seek to explain the cause of transitions.
Example of an SDSS – Predicting Typhoon Insurance Risks in China

An example of SDSS is one designed to make decisions regarding the insur-
ance risks of typhoons in China (Li et al., 2005). This SDSS is a prototype
developed by academic researchers in close consultation with industry. It il-
lustrates advanced SDSS applied to a difficult business problem. Typhoons are
a severe threat in certain areas of the world, including Guangdong Province
in China, a huge area of 70,312 square miles that borders the Pacific Ocean
for 2,105 miles. This SDSS focuses on typhoon applications for Guangdong.
The province’s prior typhoon losses were huge, involving billions of U.S.
dollars and thousands of deaths.
From an insurance standpoint, the goal of insuring against severe hazards
is to pool together a portfolio of premiums that would be unacceptable in-
dividually, but can be insured as a group. The traditional pricing approach
is “claims-based methodology,” which prices by deterministic calculations,
based on normal typhoon activity, population profiles, construction patterns,
and past insurance coverage and losses (Li et al., 2005). The problem with
this methodology is that insured firms have underestimated the loss potential
from severe meteorological hazards such as hurricanes and typhoons. One
solution is to utilize GIS combined with mathematical, statistical, pricing,
and other modeling tools. Adding the spatial component brings the model
closer to recognizing the true environmental threat.
The SDSS architecture consists of an in-house user interface and central
control module that connect to a set of modules for statistics, math, spa-
tial statistics, expert systems, and insurance pricing. As seen in Figure 3.7,
the SDSS also has a database, as well as a GIS-COM library that supports
programming features of the system. The SDSS has the capability of pro-
ducing standard business graphics and written reports. Examples are seen in
Figure 3.7 on the right.
Expert knowledge
Inference engine
Probability/year
(hazard, meteorlogy, H or R
Catalog
Structure
insurance, ...)
Library of Property
classical statistics Content
Library of Intensity
Knowledge bases CLIPS-based
Hazard Exposure
MAT LAB math engine expert system
(by its C++ compiler or H
D P/Y
AEF
MAT LAB COM Builder) The user graphic S

O
interface and Fact or instance list
central control part Intensity Loss [$]
Library of
spatial statistics Library of insurance Risk Loss
pricing modules
Z on D
Unzip
tax
GIS COM Library

Geospatial and A 1 6 1
(ArcObjects, MapObject, B 2 8 2
MAPX etc.) relational database C 5 9 3
Stochastic opt. Rate making
The Query
User's
Interfaces (Properties & Methods) Report
Specification
... ...
Relationship
GIS or standard databse management
interfaces (SQL, Oracle and Access etc.)
Figure 3.7 Architecture of SDSS for Insurance Pricing of Typhoons. Note: Object-
oriented Component Object Modeling (COM) is the interface protocol. Source: Li
et al., 2005
The SDSS modeling takes into account the following hazard and insurance
phenomena (Li et al., 2005):
1. Spatial-temporal patterns of natural hazards. Based on distributions of

the natural hazard’s past frequency, location, and intensity, correlations
can be utilized to predict future hazards and rank geographic areas on
hazard vulnerability.
2. Spatio-temporal variation of exposure. The exposure of property, peo-
ple, and facilities to hazards can be traced by the SDSS in time and
space. Total exposure can be summed up for the province, its regions,
and smaller zones (Li et al., 2005).
3. Past claims and their correlation by categories of policies. Although the
occurrence patterns of past claims are not highly reliable in predicting
vulnerability, they are a useful accessory indictor.
4. Uncertainty and other factors. The sources of uncertainty include in-
accuracies in the data, weaknesses in the models chosen, and faulty
execution of the models.
Based on these modeling principles, five models constitute the model base
of the SDSS (Li et al., 2005):
1. Hazard occurrence model. This expert systems model predicts the

spatio-temporal patterns of typhoons based on weather theories and
historical data. It creates simulated typhoons, which are stored in the
hazard database.
2. Comprehensive risk analysis model. It analyzes comprehensive risk for
entire portfolios of exposed assets based on deterministic or stochastic
simulations. Also, the model can identify geographic hot spots.
3. Zonal correlation model. It calculates spatial correlations between zones
that are vulnerable to the typhoon. This model determines insurance
rates across multiple zones.
4. Loss analysis model. The insurance rate for a zone can be computed
through a series of steps. The model computes comprehensive loss,
consisting of the direct losses from the typhoon hit, and secondary
losses from off-shoots of the typhoon itself, such as peripheral storms,
floods, and wind damage.
5. Rate making and pricing. Based on the specific exposure of an insured in
a certain zone, the model estimates the insurance rate and consequently
the premium amounts.
An example of the use of this SDSS is the dashboard seen in Figure 3.8
consisting of a map and associated graphs and tables for the fourteen regions
of Guangdong Province. “Dashboard” is a concept utilized in high-end DSS or
BI of an interactive and carefully designed display with dials, graphs, charts,
Figure 3.8 Insurance Risk Levels and Spatial Decision Analysis in Fourteen Re-
gions of Guangdong Province, China. A. Map of relative insurance risk levels of
typhoons for fourteen regions of Guangdong Province, China. B. Spatial correlation
between hazard regions. C. Wind speed intensity probability analysis for Zhanjiang
region. D. Insurance loss versus wind speed intensity for Zhanjiang region. Source:
Li et al., 2005
like a car dashboard, but in this case used for executive decision-making.
For instance, the figure shows a table of spatial correlations between hazard
regions (B in Figure 3.8), a plot of the probability of wind speed intensities
for one region in the province (C in figure), and a plot of insurance loss by
wind speed intensity for the same region (D in figure).
In short, this SDSS can be applied to complex analysis of meteorological
and hazard patterns, and insurance pricing. The model set includes statistical,
math, and expert systems models. Outputs can be mapped, graphed, or
presented in tables. Although the example is a prototype, SDSSs such as

this are implemented in some insurance firms. In terms of the chapter, this
case illustrates a custom-programmed SDSS, with strong and specialized
analysis and modeling capabilities. The benefit of a SDSS compared to a
DSS is that the digital geography of typhoons and their damage profiles for
particular land areas can be analyzed spatially, leading to improved accuracy
in estimation of hazards and risks, better decisions on pricing and premiums,
and ultimately lower cost for the customer.
SPATIAL BUSINESS INTELLIGENCE SYSTEMS
Spatial business intelligence (spatial BI) is a combination of data gathering,

data storage, and knowledge management combined with modeling, and spa-
tial analysis tools that provide complex corporate and external information
to decision makers (Gray, 2006). As mentioned earlier in the chapter, spatial
BI differs from SDSS in its greater breadth and ability to draw from more
diverse data sources. Since business intelligence is broader, it also tends to be
utilized across the enterprise (Manglik, 2006). Users are typically higher up
organizationally than for DSS, and somewhat more squeezed on their time.
A few examples of spatial BI will illustrate its somewhat broader reach.
1. Routing by Sears. GIS and BI are loosely coupled to support decisions

on managing the locations and routing of delivery personnel and repair
technicians throughout a vast network of millions of homes (ArcUser,
2005). The BI helps Sears to forecast demand for servicing and decide
on resource allocation spatially, leading to better productivity for field
personnel. The full Sears case study appears in Chapter 7.
2. In cellular telecommunications, spatial BI has been applied in Holland
to decide on the best locations of cell phone towers to maximize calling
utilization, reduce dropped cell phone calls, and to plan cell tower ca-
pacities (Gonzales, 2004). The BI user has access not only to standard
charts, tabular spreadsheet-like calculations, and thematic mapping, but
can also analyze messaging hot spots and density surfaces of users. Its
layering allows overlays of provincial and local administrative bound-
aries (Woonplats), in order to enhance decision-making involving local
jurisdictions. The BI also can overlay satellite imagery of land surfaces
and digital elevation models over cellular layers (Gonzales, 2004).
3. In real estate, current and historical property sales can be analyzed For
instance, the BI displays the proximity of sales and sales agents. This
supports decisions on the best geographic allocations of agents and
averts duplication of sales effort for one property, or zones that lack
agent coverage. Satellite imagery is provided for overlay by ESRI’s
ArcWeb Services (Gonzales, 2004).
SPATIAL BUSINESS INTELLIGENCE SYSTEMS 91
The examples demonstrate the broad reach and diverse information sources
of spatial BI.
WebFocus from Information Builders

Information Builders is an example of a leading purveyor of BI software
including the spatial BI. Standard, non-spatial WebFocus utilizes Microsoft
Excel for data manipulation and graphics. The “Spatial BI” add-on module
enhances WebFocus with access back and forth with the ArcGIS package.
Rather than combining the software into a single product, separate BI and GIS
“engines” can be connected and coordinated (Figure 3.9). A BI dashboard
displays Excel graphics, spreadsheets, and pivot tables, text files in .pdf
format, and maps from ArcGIS. The practical example in Figure 3.10 shows,
on the right of the screen, a WebFocus table listing information on high-value
insurance policies for an area in Florida having the most risk. On the left
a map shows the geographical distribution and size of these policies. The
user of the ArcGIS package has the converse benefits of being able to invoke
WebFocus modeling features, tables, and graphs.
The WebFocus and ArcGIS databases are coordinated in a seamless way.
The WebFocus database feature of accessing different databases and operat-
ing systems in a single query assist the user. The loosely coupled arrange-
ment allows the strengths of each software package to complement the other.
Figure 3.9 Overall Design of BI from Information Builders and ESRI. Note: This
Spatial BI offering combines WebFocus from Information Builders and ArcGIS from
ESRI Inc. Maps interact with Excel, .PDF, and .HTML features. Source: Information
Builders, 2006
Figure 3.10 BI Example in Insurance with Web Focus from Information Builders.
This example Uses WebFocus and ArcGIS from ESRI to locate and display the
geographical distribution of high value insurance policies in a portion of Florida who
are most exposed to risk. The user can toggle between the spreadsheet report and
map, to refine the analysis. Source: Information Builders, 2006
Another advantage is that business users can work from the BI interface they
are used to, while GIS users, typically with geo-science and environmental
backgrounds, can remain on the familiar GIS side.
A weakness of this “coordinated” approach is that the vendors have to
keep the two software packages in balance, as new versions appear and user
needs change. Also, the user must learn three packages (WebFocus, Excel,
and ArcGIS).
Map Intelligence
Map Intelligence BI software depends on Microsoft Excel for tables and
graphs, and data manipulation. It can be run with ArcGIS from ESRI or
with the desktop GIS product of MapInfo. As seen in Figure 3.10, Map
Intelligence Software is set up as a dashboard. The dashboard and map display
are synchronized (Gonzales, 2004). “Any data manipulation such as limits,
filtering, or chart slicing that are applied in the dashboard are automatically
reflected in the mapping application” (Gonzales, 2004). Data are passed back
and forth, e.g. spatial map data can be displayed on the dashboard, to highlight
trends and give sharper analytic focus.
CASE STUDY OF LARGE INSURANCE COMPANY 93
On the server side of the architecture, data are sent back and forth between
the Map Intelligence application server and a map server, commonly ESRI’s
ArcIMS. This design illustrates service-oriented architecture (SOA) running
on a small enterprise-wide internet bus, concepts introduced in Chapter 1.
It implies that Map Intelligence BI capabilities can be made available on
the web throughout the user’s enterprise, although the client components
must be installed. This BI design provides access across the wide reach of
the web.
Although the two BI examples coordinate branded BI and GIS software in
distinctive ways, both architectures allow a breadth of applications and wide
reach from diverse data sources.
CASE STUDY OF LARGE INSURANCE COMPANY
Large Insurance Company (LIC) is one of the biggest property and casualty
insurance firms in the U.S. It includes personal insurance, commercial insur-
ance involving property, casualty, and workers compensation, and a smaller
set of other insurance lines. LIC has utilized GIS software for decision-making
for over a decade, but does not have a formalized SDSS. The most signifi-
cant decision support is for commercial underwriting. It is a critical decision
function that the whole company depends on for profits, so GIS software
with modeling features provides powerful decision support. The GIS man-
ager pointed out that if GIS were removed from underwriting, the firm would
be much less responsive on short notice to underwriting requests. The critical
role of GIS for decision-making was highlighted in the two World Trade
Center bombings, for which the firm had insured clients in the buildings in
both cases.
The company has a small GIS team of seven located in Commercial Lines,
but also receives periodic loans of support people from IT such as database
analysts and a network analyst. The main applications of spatial technologies
at LIC are to manage capacity planning at property locations, support casualty
decisions, perform risk analysis to evaluate businesses for property insurance,
support underwriters in deciding on the property risks, and take actions and
manage in catastrophes. The GIS software is not currently internet- or intranet-
based, but supported through a nationwide client-server network.
In the instance of the catastrophic Hurricane Katrina in 2005, LIC had
insured in the area. Spatial analysis and mapping helped to quickly respond
to the hardest hit areas of flooding and wind damage. Data were drawn from
government and commercial sources, including FEMA, and flood maps. After
Katrina hit, the firm was able to develop a complete picture of the depth of
water throughout the metropolitan area. The disaster was of a scale that could
not be predicted ahead of time, but the catastrophe response team was able to
accurately predict damages afterwards.
LIC’s users consist of (1) underwriters and underwriter assistants, who

utilize GIS daily for location-based risk assessment; (2) the senior manage-
ment team that uses it for capacity planning and exposure analyses; (3) the
catastrophe management unit, that needs GIS for modeling of weather condi-
tions, responses to the needs of insured in catastrophes and reporting sent to
reinsurance firms, i.e. companies that accept and balance industry-wide risks;
and (4) the GIS team.
As with the typhoon insurance example, GIS feeds information into the
risk rating and premium pricing systems. Underwriters are highly dependent
on it. For example in fire insurance, the underwriter cannot price policies
without knowing the distance of a location from fire hydrants or other water
supply.
For its main spatial area of Commercial, LIC does not have formalized
SDSS or spatial BI, but rather uses GIS software enhanced with modeling of
locations and ratings. Over 1,000 commercial underwriters depend strongly
on GIS with modeling capabilities. The GIS is programmed to help the
underwriter make rule-based decisions, based on locational characteristics.
This represents the enhanced GIS approach to decision support described
earlier.
In personal lines, there is a commercial SDSS that follows simple, rule-
based logic. The basic model for personal underwriters is that the system tells
the user what he/she can or cannot offer, based on locational characteristics.
It takes a home address, geocodes it, determines the X-Y coordinates, and
then invokes a flood analysis module to determine nearby flood, earthquake,
or other hazard zones.
The two World Trade Center (WTC) bombings illustrate how far LIC has
come in its capability to support its insured in a disaster. In the original
1993 WTC bombing, the firm had insured commercial policyholders in the
tower that was bombed, but the firm could not determine using the traditional
legacy mainframe at that time, what insured businesses and people were in
the building. Instead of having a list to act on, LIC’s catastrophe group had
literally to go to the building, find out by walking around what companies
were in it, since the towers were still standing from only a garage bomb, and
check the tenants against LIC’s master list.
In the 9/11 terrorist disaster in 2001, LIC’s catastrophe team requested
from the GIS team the list of which insured parties were in the two towers.
Within minutes the GIS team produced a list of every LIC policyholder in
the building. Within hours, there was a picture of the insured’s’ losses. The
reason for the change was recognition by the CEO after the 1993 bombing
that it would be strategic to develop a complete and detailed mapping and
inventory of all insured clients throughout the country. As a result, in 9/11,
quick decisions were made and insured clients suffering huge losses were
responded to nearly immediately.
In sum, LIC has developed its national spatial decision support to the
point that it has become essential for decision-making in the underwriting
REFERENCES 95
and disaster areas, with moderate decision importance in other areas. The
dominant form of support is GIS software enabled to run models based on
data for weather, climate, insured capacity, location of clients, risk assessment,
and pricing.
CHAPTER SUMMARY
Spatial decision support has developed into an important tool for businesses.
Companies face decisions on a daily basis. Traditional tools of DSS and busi-
ness intelligence can be enhanced with spatial components, or GIS analysis
can be built up to offer robust decision modeling. The formal SDSS and
spatial BI tools are scarce in businesses, but likely to expand as they become
more affordable and user-friendly and as spatial teams in firms are better
trained. Overall the SDSS/Spatial BI area is in an early growth stage, but is
positioned to move into the subsequent stage of rapid growth. Several case
studies in banking and insurance illustrate that well-thought-out and focused
spatial decision support can yield competitive benefits. As seen by LIC, it
may take unpreparedness for a disaster to startle the firm into advancing its
spatial decision support.
REFERENCES
Agarwal, P. 2005. “Ontological Considerations in GIScience.” International Journal

of Geographical Information Science, 19(5): 501–536.
Andriendko, N.V., and G.L. Andrienko. 2001. “Intelligent Support for Geographic
Data Analysis And Decision Making In The Web.” Journal of Geographic Infor-
mation and Decision Analysis, 5(2): 115–128.
ArcUser Magazine. 2005. “Mapping Better Business Strategies: Integrating GIS and
Business Intelligence.” ArcUser Magazine, 8(3): 10–14.
Bhargava, H.K., and D.J. Power. 2001. “Decision Support Systems And Web Tech-
nologies: A Status Report.” Proceedings of Americas Conference on Information
Systems, Boston, MA, August 3–5, pp. 229–235.
Carlsson, C., and E. Turban. 2002. “DSS: Directions for The Next Decade.” Decision
Support Systems, 33(2): 105–110.
Casey, M.J., and Austin, M.A. 2002. “Semantic Web Methodologies for Spatial
Decision Support.” Proceedings of International Conference on Decision Making
and Decision Support in the Internet Age (DSIage2002), July 4–7, U.C.C, Ireland.
Crossland, M.D., B.E. Wynne, and W.C. Perkins. 1995. “Spatial Decision Support
Systems: An Overview of Technology And A Test Of Efficacy.” Decision Support
Systems, 14: 219–235.
Fonseca, F.T., M.J. Egenhofer, P. Agouris, and G. Camara. 2002. “Using Ontologies
for Integrated Geographic Information Systems.” Transactions in Geographic
Information Systems, 6(3): 231–257 .
Frank, A. 2001. “Tiers of Ontology And Consistency Constraints In Geographical In-

formation Systems.” International Journal of Geographical Information Science,
15: 667–678.
Galliers, Robert D. and A.R. Sutherland. 2003. “The Evolving Information Systems
Strategy, in Galliers, Robert D. and Dorothy E. Leidner, Strategic Information
Management: Challenges and Strategies in Managing Information Systems. 3rd
Edition, Butterworth-Heinemann, pp. 33–63.
Gonzales, Michael L. 2004. Spatial Business Intelligence. Ultimo, Australia: Integeo
Corporation.
Gonzales, Michael L. 2004. “More than Pie Charts.” Intelligent Enterprise, Novem-
ber. Available at www.intelligententerprise.com.
Goodchild, M.F. 2000. “The Current Status of GIS and Spatial Analysis.” Journal of
Geographical Systems, 2(1): 5–10.
Gorry, A., and M. Scott-Morton. 1971. “A Framework for Information Systems.”
Sloan Management Review, 13: 56–79.
New York: John Wiley and Sons.
Holland, William S. 2005. “The Fifth Dimension of GIS: Spatially Enabled Business
and Analytic Intelligence.” ArcUser, 8(3): 16–19.
Huerta, E., Navarrete, C., and Ryan, T. 2005. “GIS and Decision-Making in Business:
A Literature Review." In Pick, James B. (ed.), Geographic Information Systems
in Business. Hershey, PA: Idea Group Publishing, pp. 20–35.
Jarupathirun, Suprasith, and Fatemeh Zahedi. 2005. “GIS as Spatial Support Sys-
tems.” In Pick, James B. (Ed.), Geographic Information Systems in Business.
Hershey, PA: Idea Group Publishing, pp. 151–174.
Jarupathirun, Suprasith, and Fatemeh Zahedi. 2007. “Exploring the Influence
of Perceptual Factors In The Success of Web-Based Spatial Decision Sup-
port Systems.” Decision Support Systems. Article in press and available at
www.sciencedirect.com.
Keenan, Peter B. 2003. “Spatial Decision Support Systems.” In Mora, Manuel,
Guiseppi Forgionne, and Jatinder Gutpa (eds.), Achievements, Trends, and Chal-
lenges for the New Decade, Hershey, PA: Idea Group Publishing, pp. 28–39.
King, John Leslie, and Kenneth L. Kraemer. 1984. “Evolution and Organizational
Information Systems: An Assessment of Nolan’s Stage Model.” Communications
of the ACM, 27(5): 466–475.
Li, Lianfa, Jinfeng Wang, and Chengyi Wang. 2005. “Typhoon Insurance Pricing
with Spatial Decision Support Tools.” International Journal of Geographical
Information Science, 19(3): 363–384.
Longley, Paul A., Michael F. Goodchild, David J. Maguire, and David W. Rhind.
2005. Geographic Information Systems and Science. Hoboken, NJ: John Wiley
and Sons.
Malczewski, Jacek. 1999. GIS and Multicriteria Decision Analysis. New York: John
Wiley and Sons.
Manglik, Anupam. 2006. “Increasing BI Adoption: An Enterprise Approach.” Busi-
ness Intelligence Journal, 11(2): 44–52.
REFERENCES 97
Marakas, George. 2002. Decision Support Systems, 2nd Edition. Upper Saddle River,
NJ: Prentice Hall.
Marble, Duane F. 2006. “Who Are We? Defining the Geospatial Workforce.” Geospa-
tial Solutions, 16(5): 14–21.
Mennecke, B.E., M.D. Crossland, and B.L. Killingsworth. 2000. “Is A Map More
Than A Picture? The Role of SDSS Technology, Subject Characteristics, and
Problem Complexity On Map Reading and Problem Solving.” MIS Quarterly,
24(4): 601–629.
Murphy, Lisa D. 2005. “Geographic Information Systems: Are They Decision Sup-
port Systems?” Proceedings of 28th Hawaii International Conference on System
Sciences.
Nasairin, Syed, and David F. Birks. 2003. “DSS Implementation in the UK Retail
Organisations: A GIS Perspective.” Information and Management, 40:325–336.
Nolan, Richard. 1973. “Managing the Computer Resource: A Stage Hypothesis.”
Communications of the ACM, 16(7): 399–405.
Nolan, Richard L. 1979. “Managing the Crisis in Data Processing.” Harvard Business
Review, March/April, pp. 115–126.
Pick, James B. 2005. Geographic Information Systems in Business. Hershey, PA: Idea
Group Publishing.
Power, Daniel J. 2002. Decision Support Systems. Quorom Books.
Ray, Julian J. 2005. “A Web-Based Spatial Decision Support System Optimizes
Routes For Oversize/Overweight Vehicles In Delaware.” Decision Support Sys-
tems. Article in press. Available at www.sciencedirect.com.
Ross, Jeanne W., Peter Weill, and David Robertson. 2006. Enterprise Architecture
as Strategy: Creating a Foundation for Business Execution. Cambridge, MA:
Harvard Business School Press.
Sengupta and Bennett. 2003. “Agent-based Modeling Environment for Spatial Deci-
sion Support.” International Journal of Geographical Information Science, 17(2):
157–180.
Shim, J.P., M. Warkentin, J.F. Courtney, D.J. Power, R. Sharda, and C. Carlsson. 2002.
“Past, Present, and Future of Decision Support Technology.” Decision Support
Systems, 33(2): 111–126.
Sikder, I., and A. Gangopadhyay. 2002. “Design and Implementation of a Web-based
Collaborative Spatial Decision Support System: Organizational and Managerial
Implications.” Information Resources Management Journal, 15(4): 33–47.
Sikder, Iftikhar U., and Aryya Gangopadhyay. 2003. “Distributed Data Warehouse
for Geo-spatial Services.” Chapter 8 in Gerald Grant (ed.), ERP and Data Ware-
housing in Organizations: Issues and Challenges, Hershey, PA: IRM Press, pp.
132–145.
Sikder, Iftikhar, and Aryya Gangopadhyay. 2004. “Collaborative Decision Making in
Web-Based GIS.” Chapter 6 in Khosrow-Pour, Mehdi (ed), Collaborative Decision
Making in Web-Based GIS, Hershey, PA: Idea Group Publishing, pp. 147–162.
Smelcer, J.B., and E. Carmel. 1997. “The Effectiveness of Different Representa-
tions For Managerial Problem Solving: Comparing Tables And Maps.” Decision
Sciences, 28(2): 391–420.
Sugumaran, R., J. Meyer, and J. Davis. 2004. “A Web-Based Environmental Decision

Support System (WEDSS) for Environmental Planning and Watershed Manage-
ment.” Journal of Geographical Systems, 6:1–16.
Sugumaran, Ramanathan, Shriram Ilavajhala, and Vijayan Sugumaran. 2005. “Ex-
periences with Implementing a Spatial Decision Support System for Planning
Snow Removal Operations.” Proceedings of the 11th Americas Conference on
Information Systems, Omaha, NE, August 11–14, pp. 1668–1676.
Sugumaran, V., and R. Sugumaran. 2003. “Spatial Decision Support Systems Using
Intelligent Agents and GIS Web Services.” Americas Conference on Information
Systems. Tampa, FL, August 4–6, pp. 2481–2486.
Swink, J., and C. Speier. 2000. “Presenting Geographic Information: Effects of
Data Aggregation, Dispersion, and Users’ Spatial Orientation.” Decision Sciences,
30(1): 169–195.
Tarantilis, C.D., and C.T. Kirandoudis. 2002. “Using a Spatial Decision Support
System For Solving The Vehicle Routing Problem.” Information and Management,
39(5): 359–375.
Tomlinson, Roger. 2003. Thinking About GIS: Geographic Information Systems for
Managers. Redlands, CA: ESRI Press.
Tsou, M.H., and B.P. Buttenfield. 2002. “A Dynamic Architecture for Distributed
Geographic Information Services.” Transactions in GIS, 6(4): 355–381.
Turban, Efraim, and Jay E. Aronson. 2001. Decision Support Systems and Intellegent
Systems. 6th ed. Upper Saddle River, NJ: Prentice Hall.
Turban, Efraim. 1995. Decision Support and Expert Systems. 4th ed. Upper Saddler
River, NJ: Prentice Hall.
West, Larry. 2000. “Designing End-User Geographic Information Systems.” Journal
of End User Computing, 12:3: 14–22.
West, Larry, and Traci J. Hess. 2002. “Metadata as a Knowledge Management Tool:
Supporting Intelligent Agent And End User Access To Spatial Data.” Decision
Support Systems, 32:247–264.
Winter, Stephan. 2001. “Ontology: Buzzword or Paradigm Shift in GIS Science.”
International Journal of Geographical Information Science, 15(7): 587–590.
CHAPTER 4
ENTERPRISE APPLICATIONS
INTRODUCTION
A long-term trend for business has been towards mainstream business systems
that serve critical business functions including manufacturing, supply chain,
finance, accounting, human resources management, marketing, and sales.
They form building blocks for companies and enable processes to function
smoothly, supported by databases, data warehouses, networking, and internet
infrastructure.
In the 1990s, integrated systems appeared known as enterprise resource
planning (ERP) systems that supported groups of functions or even all of the
functions. ERPs are expensive and difficult to implement, but contribute to the
robust “back office” needed in many industries for competitive positioning.
A business system that is growing in importance is a customer relationship
management system (CRM). A CRM supports a firm in finding and com-
municating with customers, meeting their needs, and developing long-term
relationships with them. As consumers have become more discerning and
quality-conscious, companies and industries have become customer-centric.
The new way to achieve what the “friendly local merchant” used to provide is
through motivated customer-care staff who utilize customer-centric systems.
Rather than “hard sell” or “direct marketing,” the focus is on the key con-
cept of building the company-customer relationship. An example is a CRM
that supports a customer loyalty program. Prospects are tracked; committed
customers are identified and asked to join loyalty programs; and benefits are
provided to loyal customers to retain them.
99
100 ENTERPRISE APPLICATIONS
Figure 4.1 The Framework of Enterprise Applications
These are examples of enterprise applications that serve the whole orga-
nization. Figure 4.1 shows a group of enterprise applications that rely on a
technology base, shown in the center. GIS is one of this group. Alternatively,
GIS software could also be built into enterprise applications software, but so
far it is only an inherent part of some databases and data warehouses.
Enterprise systems and the central base depend in turn on the underlying
business processes. For instance, a human resource management (HRM)
system, part of an ERP, runs on a server using information stored in a database.
It is accessed by 250 users on PCs, laptops, and mobile wireless devices.
Underlying the application and technology are the HRM business processes.
For instance, four HR process steps are needed to advertise for a management
level position. The advertising is also spatial, since searches are influenced by
geographies, such as West Coast or California-only. The GIS component is
not a part of the HRM application, as seen in Figure 4.1. The GIS and HRM
system can work together through connector software. This diagram and its
concepts are returned to later in the chapter. Through their richness of data
and coordination, enterprise systems, some spatially-enabled, contribute to
the strategic competitiveness of firms.
INTRODUCTION 101
The advantages of the enterprise-wide approach are scalability, accessi-

bility, connection to external systems, ability to collaborate and cross-share
information, security, better management, and maintainability (see box).
Features of an Enterprise-Wide Approach to GIS and Spatial

Technologies
Scalability . The enterprise approach makes it easier to scale up GIS and spatial
technologies from relatively few to hundreds of thousands of users. This is
important when a company is growing rapidly or when user demand leads
to a rapid adoption increase.
Supported and accessible everywhere. In a global economy, enterprises have
extended their market reach. An enterprise approach implies that a spa-
tial application can be made widely available geographically and organ-
izationally.
Connection to external systems. Companies’ systems are becoming more col-
laborative. They are interacting with the government and nonprofit sectors.
For instance, a firm’s supply chain system involves dozens of business and
government suppliers and procurers. The firm and its providers need to
interconnect with organizations up and down the chain.
Ability internally to collaborate and cross-share information. Major functional
systems are interrelated in their business processes. Marketing projects
depend on budgetary accounting, as well as manufacturing specifications to
produce products being marketed. These imply collaborative systems that
share information.
Security . Enterprise systems tend to run on a common technology base, in-
stead of having dispersed islands of technology around the organization.
Hence the firm can focus on security and protection of the common base.
Better management. Consolidating systems into major functional modules
makes the enterprise system more understandable and manageable. The
tendency in industry to utilize well-known commercial enterprise applications
means they are easier to manage over the long term.
Maintenance. Having fewer and larger enterprise applications that are well-
known simplifies the maintenance burden over the long run.
Taken together, these features characterize the enterprise approach (Ross

et al., 2006).
The enterprise spatial perspective has only appeared in recent years in

business (Keating et al., 2003; Thurston, 2003; Sonnen and Morris, 2005).
This chapter examines the cutting edge of large, enterprise-wide spatial appli-
cations. However, 63 percent of the book’s twenty research case study firms
do not have enterprise-wide spatial applications. In fact, only one case, Rand
McNally, had widely integrated GIS and enterprise spatial applications (see
Table 4.1). Five out of twenty firms had integrated GIS and supply chain
management. The least prevalent coupling was between GIS and ERP (only
Rand McNally out of twenty firms). This may be due to the high cost and
technical difficulties in linking them up. In the sample, three-quarters of the
firms having integration of GIS and enterprise software were large ones. This
is not surprising, since large companies tend to have the resources to afford the
high cost and skills necessary to implement, manage, and maintain enterprise
software.
The heavy adoption of ERP and other enterprise applications among larger
firms encourages them to consider coupling and integrating them with GIS
software. An alternative pathway leading to GIS integration would be to
build GIS directly into enterprise software, as has already occurred with
some enterprise databases and data warehouses.
The chapter emphasizes the role of GIS and spatial technologies for en-
terprise business applications. As pointed out in Chapter 1, GIS is expanding
across the enterprise. GIS today is able to work together with and enhance the
set of enterprise business applications. The chapter first examines ERP-GIS
integration, which is illustrated by the Pidpa electrical-utility case. It then
turns to CRM-GIS integration, exemplified by the Chico’s apparel case. The
chapter discusses supply chain and its relationship to GIS, RFID, and GPS,
and then examines data warehousing and GIS. The capstone Nesa case at the
end demonstrates how a firm can integrate most of the enterprise applications
with GIS.
GIS AND ENTERPRISE RESOURCE PLANNING
ERP is an enterprise-wide, complex software application having multiple

major business applications that share a common database and/or data ware-
house. Information flows automatically throughout the ERP structure (Gray,
2006). A basic ERP is limited to key functional systems of marketing and
sales, finance and accounting, human resource management (HRM), and
manufacturing (Figure 4.2). For this ERP design, other applications such as
business intelligence and GIS, are implemented as separate software applica-
tions outside the ERP, that coordinate with it. In a comprehensive ERP, more
modules are purchased from the ERP vendor and included inside the ERP
(Figure 4.3).
The motivations for moving to ERP include the following (Gray, 2006):
r Historical. Y2K problems. At the turn of the twenty-first century, com-
panies were forced to perform software fixes to chronology bugs for the
whole enterprise. For many firms, it was easier to replace all or nearly
all their business functional systems with ERP software, rather than find
the bugs.
c04
TABLE 4.1 Case Study Results on GIS Linked to Enterprise Systems (Results shown only for the eight out of twenty case-study firms that
link GIS and enterprise systems)
Extent that GIS is linked Extent that GIS is linked Extent that GIS is linked to Extent that GIS is linked to
Case Study Firm to ERP to CRM Supply Chain Management Data Warehouse/Data Mining
Global Integrated No. No. A little. One very small A little. One focused project
Oil project in the additives used data mining and GIS.
company of GIO. Maps ESRI GIS and fuzzy logic
show suppliers, processing used to search data for
plants, customers, and upstream applications.
company locations.
October 18, 2007
Large Commercial No. No No. Yes. GIS Group extracts data

Bank on value of customers from
the data warehouse, and
16:57
geo-references it.
103
Rand McNally Yes. Rand McNally has an Yes. In large sense, GIS are Yes. GIS helps in Yes. Large data warehouses
internally developed a part of CRM. understanding parts of SCM are accessed with data
ERP. Spatial functions Plan-o-grams (shelf flows. Used for inventory mining to create input data
link the ERP with direct layouts) created for management and forecasts. that are scaled for
Char Count=
store delivery coverage customers. These linked GIS-based products.

and routes, and inventory with area demographics.
management.
Sears A little. In marketing. A little. CAMS (Customer No. No.
Allocation Management
System) allows
managers to provide
services where customer
densities are low.
(Continued)
c04
TABLE 4.1 (Continued)
Extent that GIS is linked Extent that GIS is linked Extent that GIS is linked to Extent that GIS is linked to
Case Study Firm to ERP to CRM Supply Chain Management Data Warehouse/Data Mining
Large U.S. No. No. GIS team works with No. Yes for data marts. Note: A
Property and catastrophic response data mart is smaller than a
Casualty team to give locational data warehouse and
Insurance information to specialized for what user
customers. can access.
October 18, 2007
Chico’s No. Yes. A CRM package is No. No.

used for direct mailings
and customer
16:57
information. Through
104
linkage with GIS
software, mapping and
spatial analysis supports
the CRM.
Char Count=
Baystate Health No. Slowly getting there. No. Baystate has a CRM, No. No.
but not yet been linked to
GIS.
Sperry Van Ness No. Yes. GIS and Microsoft No. No.
CRM 3.0 integrated for
commercial real estate
by adding maps and
aerials.
GIS AND ENTERPRISE RESOURCE PLANNING 105
Figure 4.2 ERP with Basic Features
r Reduced cost of commercial versus in-house software. As well-tested,

commercial ERP software became available, it was regarded as cheaper
to purchase and adapt it to the firm’s business processes, rather than to
continue a heavy custom-programming effort.
r Lack of compatibility of “silos” of separate business functional appli-
cations. Each functional silo, e.g. marketing or manufacturing software,
might work well by itself, but interfacing a silo with other business ap-
plications was too expensive. The silo problem remains a challenge for
GIS software.
ERP software from SAP, the world’s largest vendor of ERP, is present
in over 50,000 installations, with more than 10 million users in 120 nations
(Gray, 2006). Among its leading products are SAP R/3 and mySAP.com. SAP
R/3, designed to run in a client-server environment, has four core modules
Figure 4.3 ERP with Full Features
(finance, sales and distribution, human resources, and manufacturing) and

over ten optional modules. Each module supports hundreds of business pro-
cesses (Gray, 2006). SAP R/3 is designed to integrate all of the modules in
a client-server environment. It is better adapted to a hierarchical organiza-
tional structure (Gray, 2006). mySAP.com is based on internet enterprise bus
architecture with web access for users.
ERP systems are expensive, usually costing many millions of dollars. They
take several years to implement, and involve extensive training. Although em-
ployees often resist its roll-out, in the years after ERP is installed and running,
the benefits accrue, in particular greater efficiency, fewer personnel, shorter
cycle time for products to reach customers, better financial management, and
fewer people (Gray, 2006).
ERP II operates on the internet enterprise bus platform described for GIS
in Chapter 1. Based on SOA, XML, and other web protocols, it can interface
with web-enabled business software, such as CRM, supply chain, and BI.
CASE STUDY OF ERP AT PIDPA, BELGIAN WATER UTILITY 107
It also interacts with web-enabled GIS and web map services, which are
discussed in Chapter 5. The advantages of ERP II include flexibility, a wider
user base, and greater ease in open sourcing to third-party software modules.
In the ERP II environment, the user’s job is made easier by web portals,
intelligent agents, and voice recognition (Gray, 2006).
ERP and GIS software can be connected together, which takes advantage of
key strengths of each type, and yields a stronger integrated result for the user.
The ERP software is enriched by map displays and spatial analysis, while the
GIS package benefits by access to deeper and broader attribute data. ERP and
GIS software can be integrated by several alternative means (ESRI, 2006):
Remote Function Calls (RFCs). The ERP software and GIS software
invoke each other’s remotely callable functions. Calling software is
usually developed by third-party vendors.
Third-party connectors. Connectors are built by third party vendors that
directly connect packaged front-end and back-end systems. An example
is iWay Control Builder from Information Builders. This alternative,
although expensive, offers an application solution that usually provides
good performance and scalability (ESRI, 2006).
Passive middleware. ERP and GIS are connected at the level of passive
middleware, that runs on top of the operating system (ESRI, 2006). This
solution works as long as users stick to generic ERP and GIS, and don’t
try to customize their processes. An example is the SAP’s GIS Business
Connector. Both ERP vendors and third parties market these products.
Customized Enterprise Application Integration (EAI). An environment of
standards, platforms, and connector software together supports enter-
prise integration between ERP and GIS. An example is SAP Exchange
Infrastructure, which performs this comprehensive integration between
SAP and ArcGIS software.
In the list, the bottom two types are more customized and expensive, but
also involve less design and implementation cost (ESRI, 2006). In short, ERP
has a huge user base, is very expensive, and can yield improved efficien-
cies and competitiveness in back office applications. GIS, in the form of a
separate software application, can be coupled with ERP to achieve “best-of-
breed” synergies. ERP-GIS connections are undergoing rapid development,
and nearly complete integration might be achieved in the future. The Pidpa
case demonstrates the synergies of well-conceived ERP-GIS integration.
CASE STUDY OF ERP AT PIDPA, BELGIAN WATER UTILITY
Pidpa, a Belgian water utility firm based in Antwerp, is among the largest
drinking water firms in the Flanders portion of the country. It supplies
Figure 4.4 Pidpa’s Water Distribution Locations in Antwerp Province, Belgium.

Source: Reynaert, 2005
drinking water throughout Antwerp Province. Figure 4.4 shows the firm’s wa-
ter production centers and distribution network. Pidpa’s water comes mostly
from groundwater, which is sometimes polluted adding to the firm’s empha-
sis on water quality. Pidpa samples the water quality at a multiple network
points. The sampling is supported by a lab information management system
(LIM) that does continual quality checking. The Scada (Supervisory Control
and Data Acquisition) system automatically controls its water distribution
network. Pidpa also has hydraulic modeling software, and AutoCAD-based
Synoptiek software that allows for designing detailed sketches on its water
distribution network. The network has more than 7,500 miles of water mains
and produces over 72 million cubic meters of water yearly.
Pidpa adopted a full-featured implementation of SAP/R3 ERP, shown in
Figure 4.5, which includes SAP’s four standard modules, plus others for
control, asset management, funds management, project systems, and plant
maintenance. The firm adopted the ArcGIS family of products, combined
with ArcFM, a spatial module from Miner&Miner/Talvent for the engineer-
ing management of utilities. The GIS database is managed by ESRI’s data
management software, ArcSDE. The web-server software ArcIMS supports
web and mobile users.
Pidpa performed in-house programming for its web viewer and for its SAP-
to-GIS connection. It commissioned ESRI Nederland B.V. to custom-program
GeoLink, a web viewer for the GIS database based on ESRI’s ArcIMS HTML
viewer. It provides web viewing of maps and data from the SAP ERP, Scada,
and CIS, a customer information system. Costs to develop GeoLink were
limited due to the presence of a solid GIS and SAP architeiture. A benefit to
CASE STUDY OF ERP AT PIDPA, BELGIAN WATER UTILITY 109
Figure 4.5 Pidpa’s ERP System. Source: Reynaert, 2005
users is that GeoLink can act as a web portal to access diverse information
relevant to Pidpa’s customers and employees. In sum, Pidpa’s GIS system,
seen in Figure 4.6, is largely standard with the exception of the GeoLink web
browser.
Although GeoLink solved the need for web viewing, a custom approach
was adopted for integration of SAP and GIS data. Pidpa custom-programmed
its own “third party connector,” but it limited the integration to eight business
scenarios that users rated as essential.
One scenario in the GIS-ERP connection is “calamity point,” which Pidpa
calls a water-utility maintenance problem that is not quickly fixable, but rather
long-term in nature. Calamity points are marked and accumulate over time.
They serve a strategic purpose for management, when it plans for renewal of
major segments of the network (Reynaert et al., 2004). A set of calamity points
is shown along a water distribution line in the top of Figure 4.7. The user can
highlight a calamity point as seen in dark shading at the top of the map and re-
quest the software to provide automatically from SAP’s maintenance module
a listing of the service orders for fixing it, which are shown at the bottom.
As seen in Figure 4.8, besides invoking SAP tables, the GeoLink web
browser can also automatically call up Scada summary tables, CRM infor-
mation, and scanned water-network schematic drawings from Synoptiek.
In summary, Pidpa succeeded in implementing a full-featured ERP,
GIS, CRM, and commercial Scada utility software, along with GeoLink, a
Figure 4.6 Pidpa’s GIS System Design. Source: Reynaert, 2004
customized web map browser. It is designed so users can click on map fea-
tures and gain access to information from the other enterprise business and
utility systems. Pidpa plans to modify the GeoLink map browser to enable
use by field workers who have mobile devices. Further, it hopes to increase
the extent and power of integration between GIS and the other systems.
GIS AND CUSTOMER RELATIONSHIP MANAGEMENT
Customer Relationship Management (CRM) refers to a business strategy or

application intended to improve customer satisfaction and in turn to grow rev-
enues and profits (Oracle, 2006). CRM also encompasses software packages
to achieve this and the transformation of an organization through new thinking
about customers (Oracle, 2006; Gray, 2006; Fjermestad and Romano, 2006).
Customers who sometimes feel neglected benefit by the personal attention
and customized services provided by CRM. Direct, personal interactions
between the company and customer, termed “touch points,” build and rein-
force the customer relationship (Oracle, 2006; Gray, 2006). CRM helps in this
process by streamlining targeted information and providing it to the customer
GIS AND CUSTOMER RELATIONSHIP MANAGEMENT 111
Figure 4.7 Pidpa: User Selection of Calamity Point in GeoLink Browser Showing
Link to SAP Work Orders. Source: Reynaert, 2005
Figure 4.8 Pidpa’s Links of GIS with Major Business Systems. Source: Reynaert
et al., 2004
and customer-service employees. Most CRMs have business intelligence fea-

tures, so analysis and modeling can be done to support better decisions for
the customer.
CRM emphasizes the “best” customers, i.e. those most remunerative to the
firm. It’s a rule of thumb in marketing that a fifth of the customers provide
four-fifths of the profits (Gray, 2006). Helped by CRM support, customers
gain loyalty that can be reinforced for years or a lifetime. The benefits of
CRM to the company are to: (1) identify good customers, (2) develop greater
customer satisfaction and loyalty, (3) close more deals more quickly, and (4)
improve market segmenting, which results in better targeting of customers
(Oracle, 2006; Gray, 2006). Understanding the location of customers, facili-
ties, employees, and assets can enhance the benefits.
Specific functions of CRM are to improve customer service (e.g. give call
center and customer agents much more information on customer history)
and strengthen personalized marketing by the sales force, i.e. sometimes
referred to as sales force automation. This encompasses design of marketing
campaigns and capture of information on customers, sources, and regional
demographics (Oracle, 2006).
CRM software is available from vendors that include Oracle’s Siebel
Systems, M2M’s Onyx Software Corporation, and Microsoft. Additionally,
CRM is offered as a module in most ERP products, such as SAP’s R/3 and
GIS AND CUSTOMER RELATIONSHIP MANAGEMENT 113
mySAP.com. Oracle is a leading, full-featured vendor that offers client-server

and hosted web services products through its Siebel products. M2M’s Onyx
is a medium-sized CRM vendor that emphasizes professional markets. CRM
software has the following features: (1) product management, (2) sales ca-
pability in the field and on the web, (3) collaboration (with partners and
internally), (4) marketing automation, (5) customer service, and (6) business
intelligence (Gray, 2006). BI, covered in the last chapter, provides analysis
and modeling to help managers cultivate and handle customers better.
CRM software, only a decade or so old, is still in a low maturity stage,
and has tended to have a high failure rate, as much as 50 percent or higher
according to the Gartner Group (Gray, 2006). To overcome these problems,
CEO Janice Anderson of Onyx (Financial Services Technology, 2006) points
to four key ways to achieve success with CRM: (1) Develop customer in-
formation in the CRM that can support many business units and products.
This enables diverse sales people to cross-sell products to a customer. (2)
Customize the customer’s solution to his/her individual needs. The customer
does not want a solution that replicates every competitor’s CRM. (3) Adapt
the information and uses of the CRM to changes in markets and customers.
CRM design and implementation might take significant time, during which
customer needs change rapidly, so when rolled out the solution might appear
old-fashioned. (4) Integrate the CRM in real-time with other major business
systems. This comes back to the chapter’s enterprise framework in Figure
4.1, which shows the integration of CRM with eight or more major business
application modules (Financial Services Technology, 2006). Another mis-
take is that a company can rush to implement CRM without developing its
customer strategy (Gray, 2006). Furthermore, privacy issues with CRM can
develop when personal information is used for purposes other than originally
intended, without appropriate permission from the customer. An example in
a spatial context is that customers and their personal information might be
identified through easy recognition of their mapped locations. Privacy issues
are covered in Chapter 11.
Spatial systems enhance CRM functions by recognizing location. Spatial
analysis refines benefits of CRM to the customer as follows (Winslow and
Lea, 2002):
Data collection and enhancement. GIS can be helps in error-cleaning cus-

tomer data. The process of geocoding addresses and mapping customer
data can reveal errors in the data. For instance, a map shows the same
customer at multiple principal residences or it displays many headquar-
ters for a single business customer. These can be checked and corrected,
on the ground if necessary. Spatial analysis can be useful in imputation
of missing values. If a customer has missing attributes, certain values can
be estimated by taking averages of customers who are geographically
adjacent. In a different setting, imputation of missing values is done for
small areas by the U.S. Census.
Business intelligence. BI is useful in CRM for data mining, modeling,

and forecasting. As indicated in Chapter 3, many BI techniques can
be spatially enhanced. Data mining is a method for looking for certain
characteristics or data relationships in a large-scale dataset, a process
strengthened by including spatial location of entities. Forecasting is
often strengthened through spatial statistics and spatial econometrics.
For instance, the error in regression forecasting models for housing
prices can be reduced by including location (Wachter et al., 2005).
In the telecommunications industry, spatial BI can sustain customer rela-

tionships by reducing cell phone churn (Winslow and Lea, 2002). Cell phone
markets suffer from rapid turnover of customers who are disappointed by
dropped calls, poor service, and malfunctioning equipment. Users come to
depend on always having good outdoor reception regardless of location. Us-
ing GIS and data from GPS-enabled customers, spatial BI can analyze how
the customer’s locations are correlated with his/her demographics, the time
of day of use, and call volume in the area (Winslow and Lea, 2002). “By
analyzing the customer’s locations, the risk of losing a particular customer
can be predicted months ahead” (Winslow and Lea, 2002). As a consequence,
attention can be given to checking and correcting that customer’s usage prob-
lems, and special customer service can be applied such as offers of new cell
phone plans or providing promotional maps showing the growth of existing
cell-service areas.
Site Evaluation Models

Corporate customers in many industries can be assisted by site evaluation
models. Although the models can be run non-spatially, they are improved
through GIS. In the real estate industry, customers are interested in visualizing
the siting aspects of their properties, as seen in the first chapter’s Sperry Van
Ness case study. In banking, site evaluation can be done of branch locations,
such as the last chapter’s LPCB case, or real estate site evaluation services
can be provided for high-value customers. Spatial analysis can model optimal
siting of business facilities, utility lines, and transport corridors.
Distribution of Resources
The workforce and investments being applied to enhance customer relation-
ships can be modeled spatially. For instance, sales force automation seeks
to allocate a sales force in the best way to identify customers and develop
relationships. Map layers of the locations of the sales force and its customers
can be overlaid and compared. Misallocations can be corrected. Such overlays
are useful also in call centers in the telecommunications industry. Call-center
operators can visualize the location of customers phoning in, relative to nearby
service facilities (ESRI, 2006).
CHICO’S CASE STUDY: GIS AND CRM IN A CUSTOMER-CENTRIC FASHION COMPANY 115
To accomplish these benefits, CRM and GIS interact as separate software

packages. GIS and CRM are tied together by software connectors developed
in-house or by third-party vendors. For example, Paz Oil, an Israeli consumer
oil and gasoline firm, implemented SAP’s CRM module connected to ESRI
ArcGIS software. Paz has 260 filling stations that also include convenience
stores. Its subsidiary Pazomat provides fuel to customers that have fleets
of vehicles. The fleets have gas-tank-mounted devices that read the fuel
levels and transmit the information to Pazomat’s IT system. The GIS and
SAP’s CRM module are linked together through the integration hub of SAP’s
NetWeaver XI (SAP, 2006). Pazomat can view maps that display the locations
and fuel levels of customers’ fleet vehicles, allowing Pazomat to adapt to the
spatial distribution of customer gasoline demand throughout its network.
The interaction of GIS and IT with CRM can be examined through a two-
way matrix that has the process stages of customer relationship as columns
and factors of CRM as rows (Gray, 2006). The four stages are (1) identifying
prospective customers, (2) differentiating customers by their values and needs,
(3) interacting with the customer, and (4) customizing a response. As seen in
Table 4.2, modified from Gray (2006), the process stages are linked to CRM
sub-factors and corresponding IT and GIS uses.
For instance, identifying customers can be supported by cookies, cus-
tomized websites, and product personalization. At the end stage of cus-
tomization, with CRM fully implemented, the IT “heavy-hitters” of ERP
and e-commerce can be brought into play to deliver maximal information
both to customers and their relationship managers.
The matrix has been modified to include GIS shown in the bottom row.
In the identification stage, GIS can provide personalized maps to customers,
while in the differentiation stage, spatial BI can be applied to refine customer
groupings. For instance, cluster analysis results can be mapped to distinguish
customer groups. In the interaction stage, web and wireless interactions can
be enhanced through maps and analysis. Maps can be made available to
customers on their cell phones to assist them with a retail service. Finally in
the customization build-out at the end GIS can be coupled with ERP, CRM,
and other powerful enterprise systems.
In summary, CRM is an approach to give attention and personalized interest
to customers. GIS helps CRM better realize this relationship building by
incorporating location and spatial processes.
CHICO’S CASE STUDY: GIS AND CRM IN A CUSTOMER-CENTRIC

FASHION COMPANY
Chico’s is a women’s apparel chain that emphasizes customer service and

appeals to a “mature” (forty-five-year-plus) market. Founded in 1983, by
2006 Chico’s had store, catalog, and web sales that totaled $1.4 billion,
and employed 11,000 persons. Its rapid growth is shown by addition of
c04
TABLE 4.2 GIS and IT Factors in CRM
Process Factor Identification Differentiation Interaction Customization
CRM goal Identify individual Evaluate customer value Build a continuing Fulfill customer needs,
customer and needs relationship generate profit
Traditional mass Not done Clustering Call center Sales, services
October 18, 2007
marketing
CRM sub-function Customer profiling Individual level analysis Call center management Sales force automation,
for individual. marketing
16:57
Customized auto response
116
IT used Cookies, website, Data mining, Web applications, ERP, e-commerce
personalization organizational learning wireless
communication
GIS used Personalized mapping Spatial BI Spatial web, spatial GIS and spatial
Char Count=
wireless technologies (GPS,

RFID) seamlessly
linked to CRM, ERP,
and e-commerce
Gray, Paul. 2006. Manager’s Guide to Making Decisions about Information Systems. New York: John Wiley and Sons.
CHICO’S CASE STUDY: GIS AND CRM IN A CUSTOMER-CENTRIC FASHION COMPANY 117
150 stores in 2007 (Chico’s, 2006). It has always emphasized customer loy-
alty and direct marketing (Roussel-Dupré, 2002). This is highlighted by its
Passport Club which requires $500 in cumulative purchases for membership.
There are 1.7 million permanent members of the Passport Club and 334,000
members for its less expensive White House/Black Market chain of stores
(for thirty-five-year-plus age group). The members of the clubs provide 80
percent of Chico’s revenues. Members’ Passport credit cards provide person-
alized service, discounts, coupons, and birthday gifts. The Passport customer
gives detailed information in her application that includes home address.
Subsequently her home location is geocoded. Geocoding is the process of
applying GIS software to convert street addresses into X-Y coordination lo-
cations. Hence, the customer home locations appear as points on a map layer.
The purchase information of customers from all the stores is uploaded
weekly from the store’s point-of-sale databases and consolidated into a na-
tional database that is provided to the GIS team to analyze. It includes each
Passport customer’s residence and principal store location, demographic in-
formation, purchase history, and the socioeconomic profile of her home small
geographic area (U.S. Census block group).
Chico’s sales approach is characterized by sales personnel who offer an
attentive and personalized approach to customer care. Typical customers
demand new apparel frequently so there are rapid inventory turns. The phi-
losophy is that employees act as if they work for a small local store, e-mailing
customers, being friendly to customers, and even calling them by first names.
Chico’s has capitalized on its loyal customer base by implementing a suc-
cessful CRM supported by GIS. These systems enhance responsiveness to
customers and improve Chico’s employees’ knowledge of them.
The CRM and GIS systems also support real estate analysis of new and
existing store sites. The CRM information is used for decision-making on
opening, closing, or changing store locations. The AnySite software from
MapInfo can analyze the relationship between the current store location pref-
erences of customers and prospective real estate locations. For instance, if a
new store site is being evaluated, customers can be mapped color-coded by
their primary store preference. If an area reveals a mixture of colors with no
clear store preference, that points towards the need to locate a new store in that
area. The map also shows topographic and man-made barriers such as rivers,
highways, and hills that may influence whether or not, and where a new store
is designated. The amount of “cannibalization” (i.e. a new store taking away
sales from old ones) can be predicted, and steps taken to adjust to the correct
amounts of merchandise at each outlet. It can even make adjustments, so that
employees at an old store are not unduly deprived of bonuses, once the new
store arrives and draws away business.
Chico’s MIS system is housed in central, large IBM servers at its Fort
Myers, Florida, headquarters and Georgia distribution center, that connect
with decentralized application servers located at store sites. The company does
not yet use an ERP system, but one is planned for rollout in 2008. Instead there
are a group of specialized application packages, many leading ones for the
retail industry (Chico’s, 2006). For CRM, Chico’s uses the Connected Retailer
from NSB, which supports CRM as well as store merchandising, planning,
allocation/replenishment, and sourcing. MapInfo Professional GIS software
is run alongside the Connected Retailer and utilizes the same database of
customer information.
The CRM and GIS support a number of key features that relate to Chico’s
customer-centric focus.
Direct mailing. An estimated 5 million items are sent monthly to cus-
tomers, including event promotions, coupons, and catalogs (Roussel-
Dupré, 2002). The CRM refines this mailing through analysis that gives
the optimal customer audience for a particular mailing.
Unified customer database. Prior to the CRM, each sales channel had
its own customer information system. The CRM gathered them into a
uniform customer database that supports cross-organizational flows of
information (Roussel-Dupré, 2002).
Business intelligence for CRM. Chico’s was able to perform analytics to
make customer relationship processes more efficient and to understand
customer patterns better. For instance, Chico’s was able to determine
that its best customers on the average shopped in a Chico’s store every
four to five weeks. Through CRM and GIS, it was able to find out
where customers shopped and what they bought at particular locations.
In Florida, Chico’s many stores have a seasonal customer flow. A woman
Passport member from Chicago vacationing in Florida may purchase
two tops in Florida, a transaction that can be compared with her purchase
profile back home.
Forecasting. The CRM enables the firm to predict, based on historical
records, how the customers residing in an area would respond to a sales
promotion. GIS is used to map the results.
Chico’s CRM/GIS initiative has worked. Before the CRM, Passport mem-
bers represented only 9 percent of Chico’s revenues, while three years after
implementation, they accounted for 80 percent. During each store visit, a Pass-
port member spends 40 percent more than a non-Passport/member (Roussel-
Dupré, 2002). GIS has reinforced CRM, by taking into account where cus-
tomers shop, what they buy where, and what is the geography of customer
relationships. The strategic success of CRM coupled with GIS is tied to its
synchrony to Chico’s key value of developing and sustaining customer loyalty.
GIS AND SUPPLY CHAIN MANAGEMENT
Supply chain is the flow of goods and services in an organization. Goods orig-
inate in raw materials, are transported, assembled into components, further
GIS AND SUPPLY CHAIN MANAGEMENT 119
assembled into finished products, distributed, sold, and sometimes returned.

Supply chain management uses modeling and analysis techniques to improve
performance and functioning of the corporate supply chain (Gray, 2006).
Information systems including GIS are important in controlling and coordi-
nating the supply chain, and enhancing its efficiency and profitability.
The factors in supply chain management (Gray, 2006) are:
1. Planning. The products or services must be designed, and the multiple

steps planned out to end up with the finished result. The IS to support the
supply chain must be developed to enhance communication, efficiency,
and accuracy. Along with this, the GIS and spatial technologies must be
planned.
2. Sourcing. Sources for raw materials and components need to be deter-
mined, and optimized for cost, shipping distance, and availability.
3. Manufacturing/creating. How will the components be manufactured
into a product? The process of component assembly needs to be deter-
mined.
4. Delivering. The approach for the final delivery of products to customers
must be worked out. This includes whether the product will be deliv-
ered pre-assembled or as parts. Delivery must be coordinated with the
customer’s schedule.
5. Returning. If a product is not acceptable to the customer, return of
the product to an earlier step in the supply chain must be planned. A
returned product rarely can be sent back along the same forward route
from which it originated. The returned items are in smaller quantity,
and usually go back via different geographic routes. Often return is
done through a third party, that can wait until a sufficient lot size is
accumulated to cost-effectively reverse-ship the items.
IT supports information flows that inform, analyze, and optimize the phys-
ical flows. IT for “planning” (Factor 1 above) consists of determining the
end demand for the finished product, and then working backwards to esti-
mate the demands upstream in the chain. A second type of software program,
“implementation,” supports Factors 2 through 5. The focus is to reduce the
length of time for manufacture and deliveries and make adjustments so de-
mand targets are met at each step. Often there are multiple suppliers in
the chain, that lead to poor estimates, misunderstandings, and underperfor-
mance. If the supply chain is internal to a firm, the same problems can
occur, although there is the advantage of being under a single management
structure.
Supply chain has a close relationship with levels of inventory (Gray, 2006).
If demand forecasts lead to overproduction, the surplus might have to be
abandoned, given away, or written off, while underproduction in the chain
results in disgruntled customers.
Supply chain software is available as modules in ERP packages such as

SAP. The ERP approach has the advantage of tight, seamless exchange of
supply chain with the other major business systems, but the minus of reduced
flexibility in modifying the supply chain parameters for a particular client.
Supply chain is also available as standalone software from such vendors as
Logility Inc., Catalyst International, Aspen Technology, and Clarus Corpo-
ration. Supply chain software can reside inside or outside of ERP (compare
Figures 4.2 and 4.3), and can consist of commercial or in-house programs.
The problems and challenges for supply chain success are (1) trust among
suppliers, manufacturers, distributors, customers, and information vendors;
(2) employee resistance (supply chain often involves displacement and loss
of jobs); (3) lack of training in the start-up phase (supply chains rarely run
“out-of-the-box,” they need tuning, adjustments by parties involved, and re-
tuning); (4) consumer preferences that lead to new product designs which
imply modified supply chain; and (5) errors in forecasting demand. Demand
fluctuates depending on environmental conditions, political regimes, trans-
portation strikes and delays, economic ups and downs, and other factors, so
it rarely can be accurately estimated (Gray, 2006).
GIS and spatial technologies contribute to more efficient and faster supply
chains. Any item moving in the supply chain with recorded geographic coor-
dinates can be more accurately tracked through a combination of GIS, RFID,
and GPS. Examples of how GIS contributes are as follows:
Spatial analysis of supply, manufacture, and delivery points. Transport

and shipping durations can be better modeled by incorporating geog-
raphy (ESRI, 2006). Consider a supply chain in a region impacted by
an earthquake. Vehicle transportation is halted or slowed due to physi-
cal barriers and dangers. Shipping is impacted due to harbor and dock
damages. Airports are temporarily closed. However, zones vary con-
siderably in their extent of transportation disruption. Spatial analysis
can estimate the geographical pattern of damages and delays for dif-
ferent transport modes, and can include location to better estimate how
supply-chain transport timing will be altered.
Spatial distribution of customers in the supply chain. Customers are the
source of supply chain demand which is hard to predict. The present
and future locations of end customers who receive goods are a critical
aspect of demand. Their geographical prevalence in sales areas helps to
determine the allocation and timing of the final supply routes to sales
locations.
RFID and GPS can receive input data on the spatial locations of supply-
chain items and transport vehicles, which are combined with attribute data
and utilized by GIS to perform spatial analysis. RFID-tagged items are read
as they pass fixed reader locations in the supply chain. An ID on the tags
allows records to be accessed on a remote server. If the readers are spatially
GIS AND SUPPLY CHAIN MANAGEMENT 121
referenced, mapping can be performed. Businesses are beginning to make use

of RFID, but the proportion going further to apply GIS is presently small.
The example of the U.S. military points to growing use. The military is
the leading organization using RFID and GPS in supply chain. It imposed
RFID requirements for all its suppliers by 2005. It demonstrated the wartime
strategic value of these technologies in the Iraq War where the military utilized
RFID and GPS to input location data into real-time GIS to assist in battlefield
decisions (Galuszka, 2005).
RFID was introduced in Chapter 1, with the Wal-Mart example. Although
the focus of this book is on the private sector, the military use of spatial supply
chain is included, since it is at the leading edge and points to where many
businesses are heading. However, there should be caution that not everything
from the military carries over to business.
The U.S. military’s huge and complex supply chain includes 43,000 sup-
pliers. Correct inventory must be supplied to hundreds of thousands of per-
sonnel, and life and death might depend on it. For instance, a combat unit
that is quickly running out of ammunition and food must have delivery at a
particular place by a given time.
The military spent over $100 million to implement RFID as a standard by
2005, as a part of Secretary Rumsfeld’s modernization of military technology
(Galuszka, 2003). In the Iraq War, the U.S. military was the first one to use
RFID for supply chain in combat. Military commanders were able to monitor
the supply-chain flow of equipment and supplies in real time (Military Medical
Technology, 2005). RFID tags were affixed variously at the individual item,
box, carton, and pallet levels.
For higher-value ordnance, active tags were used at the item level that
can be read by fixed RFID readers within a 150-foot radius. All shipping
containers have active tags that indicate in detail what is inside, and can
be read by military satellites that pass by every 6 to 12 hours. Hence, all the
supply-chain inventory having active tags can be viewed and analyzed. In Iraq,
hardened RFID readers, spaced at 120 guarded locations, collect information
on the contents of passing supply convoys instantaneously (Galuszka, 2005).
At U.S. Central Command (CENTCOM) in Tampa, GIS maps of the war
zone show the shipping containers and high-value items, as well as aircraft,
drones, vehicles, and fighting action on the ground. When an item is moved
in a vehicle, GPS can track the vehicle and its contents in real-time.
The benefits to the military are multiple. (1) In a dynamic battle situation,
supplies can be moved to the right place quickly. (2) Cargo can be dropped
or sequestered in a “secret” location and picked up later. (3) What is typically
a “just-in-time” supply chain can be changed to “sense and respond,” which
refers to flexibility to change the supply chain while operations are underway
(Galuszka, 2005). For example, a combat unit under pressure could escape
with limited gear, while leaving equipment and RFID-tagged supplies behind,
which are tracked by satellite to be picked up later. Hence, receiving supplies at
an exact time and location becomes somewhat less important. (4) The supply
chain contents become “visible” meaning that the contents of all the items
in the chain are visible system-wide without opening packages, containers,
or pallets (Sun Microsystems, 2006). (5) Inventory management is improved,
which leads to reduced supplier fraud and pricing errors, and better theft
detection and control of returns.
This advanced military system does not eliminate some problems. Predict-
ing demand remains difficult, with the risks of over- or under-production. In
the Iraq War, there was initially severe shortages of body armor and armored
hummers. There were also problems in faulty RFID readers, inaccuracies
of reading, and gaps in satellite reception. Nevertheless, this military supply
chain based on RFID and GPS has been successful in and out of combat.
GIS is coupled with RFID and GPS to provide visualization and spatial
analysis.
Although Wal-Mart as an industry leader trails the military, its require-
ments set in 2006 of having its largest suppliers attach RFID to shipping
crates and pallets supports similar potential to the military, but under dif-
ferent operational constraints. There is a leveraged effect for suppliers to
adopt RFID upstream and downstream in this huge supply chain, and other
major retailers are following Wal-Mart’s example. It will be helped by the
anticipated drop in pricing of passive RFID tags from 25 cents to under 1
cent based on the expanded volumes expected within a few years (Williams,
2004).
In summary, the supply chain represents a major business system. As items
move geographically through it, location information can be detected by RFID
and GPS and spatially analyzed through GIS. The IT control and management
of the chains is made more accurate, while they can also be more quickly
altered and reconfigured. The U.S. military led the way in achieving a cutting
edge spatially-enabled and combat-ready supply chain, while Wal-Mart and
other leading companies are not too far behind.
SPATIAL DATA WAREHOUSES AND DATA MINING
The data warehouse is an alternative form of data storage from the con-
ventional operational database. It is oriented towards a view of data that is
subject-oriented, rather than application-oriented. Subject-orientation refers
to objects that represent the key elements of decision-making in business,
such as customers, markets, products, and employees (Laudon and Laudon,
2006). By contrast, the traditional operational database focuses on the func-
tional areas of the business, such as sales, finance, and manufacturing. An-
other feature is that the data warehouse retains older data so it can look
at change tendencies over time. The operational database regularly purges
older data for deletion or archiving. Since the philosophy is to store every-
thing, the size of storage can be potentially huge, a true “warehouse.” The
SPATIAL DATA WAREHOUSES AND DATA MINING 123
Figure 4.9 The Data Warehouse and Its Data Flows. Source: Gray, 2006
data are nonvolatile, i.e. after being stored, they remain the same over time.
This lends stability to the data in a data warehouse (Inmon, 1992). The data
warehouse concept also favors summarized data, which are useful in decision-
making and also become nonvolatile. Granularity distinguishes the individual
data items from the summaries, i.e. the most granular refers to raw data,
while less granular is summarized data (Gray, 2006). An example of sum-
marized data is a summary of account totals for March, 2007, for a business
department.
At certain time points, data are moved from operational databases into the
data warehouse for permanent, long-term storage. In a large organization,
often data are input from a wide collection of databases on the inside and
outside. The process of error-checking and entering these data is known as
ETL (extraction, transformation, and load). As shown in Figure 4.9, data enter
the data warehouse from the operational databases. Once inside, they can be
accessed by a variety of analysis, statistical, data mining, and GIS software
(Gray, 2006; Laudon and Laudon, 2006). Also shown are metadata, which
keep track of the detailed descriptions of the records and entities in the data
warehouse.
Data warehouses have been particularly useful for the applications of data
mining, direct marketing, online analytical processing (OLAP), business in-
telligence, CRM, and supply chain management.
Data mining seeks answers to business problems directly from the raw
and summarized data and from data relationships in the data warehouse.
The answers are not pre-determined but often discovered through exploratory
methods (Gray, 2006). A variety of data mining techniques are utilized which
are beyond this book’s scope: intelligent agents, expert systems, fuzzy logic,
neural networks, exploratory data analysis, and data visualization (Tan et al.,
2005; Gray, 2006). All these methods are able to intensively explore large
amounts data and/or relationships, and to identify potential answers to com-
plex business problems.
Direct marketing uses techniques that segment customer data based on
market distributions. They are examined further in Chapter 5. Online An-
alytical Processing (OLAP) is a set of rules for accessing and process-
ing multidimensional data. OLAP rules, covered in Chapter 8, enable
accessing data based on simple and efficient steps rather than complex
models.
Data warehousing and GIS form a strong analytical combination. The GIS
provides spatial analysis and visualization of geo-referenced data in the data
warehouse. For instance, it can enrich the uses of data warehouse information
for utilities infrastructure; energy exploration, production, and distribution;
traffic accident analysis; large scale auto insurance risk analysis; management
of fleets of vehicles; and business intelligence for decision-making (SQL
Server Magazine, 2002; Farley, 2004; Location Intelligence, 2006).
For data warehouses in auto risk insurance, maps can be produced that take
spatial views from the usual ZIP code geography down to hundreds of small
areas within the ZIPs (Location Intelligence, 2006). This allows underwriters
to set more refined policy pricing. Another example is the City of Portland’s
spatial data warehouse for city and regional traffic accident data which has
fourteen dimensions and sixteen years of data. As a result, users of the city’s
system have access to interactive mapping of all the accident locations during
a decade and half, which accompany accident reports and give added insight
for decisions (SQL Server Magazine, 2002).
GIS and spatial analysis can be in use for data warehouses at various
steps along the data flows shown in Figure 4.9, a topic of Chapter 8. Data
warehouses, which constitute an important part of enterprise architecture,
can be spatially-enabled and have proven valuable for many larger-sized
firms.
CASE STUDY: NESA, DANISH ELECTRICAL UTILITY
Nesa, a medium-sized Danish electrical utility firm, has not only installed a
group of enterprise systems including a GIS; an ERP system; and Scada, an
enterprise-wide process control system; but it has implemented SmartLight,
which allows consumers to automatically control outdoor lighting. Coming
at the end of the chapter, the Nesa case demonstrates an enterprise philosophy
and approach.
Nesa has over one million residential, business, and government customers
and a large share of the Danish electrical utility market. Its IT mission is
CASE STUDY: NESA, DANISH ELECTRICAL UTILITY 125
to “support and improve the competitive power and the business basis of
Nesa” (Nesa Annual Report, 2006). To achieve the mission, Nesa decided to
focus on leading software packages to support major business applications,
rather than on lesser-known software from smaller vendors or on in-house
applications. Its large vendors include SAP for ERP, ESRI for GIS, Siemens
from CRM, and MDSI for management of mobile workforce. Through an
enterprise approach, Nesa’s goal is to achieve a high extent of internal and
external integration. It hopes that the software consistency and standards can
lead to better customer service in concent with its vision to meet customer
needs and create value and relationships with them (Nesa, 2006).
A general schematic of Nesa’s IT application configuration rests on a
bottom level of communications and process control (see Figure 4.10). At
the lower right, the Scada process control refers to a type of utility soft-
ware that provides integrated monitoring and control of the entire electrical
utility network. For instance, if a part of an electric transmission line fails,
Scada registers the failure, assesses what the impacts are, takes automatic
corrective action, and notifies managers about the problem and its potential
impacts.
At the middle level in Figure 4.10 is the ERP back-end system that includes
the modules of material management, payroll, maintenance and service,
project management, finance, and control (Engsbro and Westmose, 2005).
The ERP system and its modules run on top of a data warehouse. Along-
side the ERP are separate GIS software packages (ArcGIS and ArcFM) and
computer-aided design software (Auto CAD and Grid Design). Arto CAD
software provides computer-generated drawings for engineering design, e.g.
design of a utility transformer station or a power plant. It is linked through a
standard interface with GIS.
At the top of the figure are software applications close to the customer, in-
cluding financial applications of trading and portfolio analysis, management
of the mobile workforce (MDSI and SRMON), telephone and voice systems,
the SAP CRM module, printing and distribution, web software, SmartRead
software for automated meter reading, and SmartLight for automated lighting.
SmartRead automatically reads and transmits customer utility meter informa-
tion into the SAP Customer and Maintenance modules. As seen in Figure
4.11, the SmartRead meters at the customer premises feed this information
to SmartRead monitors through wireless communications. This innovation
saves the firm the high costs and inaccuracies of regular meter reading by
service technicians. Rather, a technician visits the meter only if it needs to
be maintained. For the SmartRead system, GIS displays the design maps of
network transmission flows.
Nesa has been innovative in designing and implementing a system called
SmartLight (Engsbro and Westmose, 2005) (see Figure 4.12) for automated
monitoring and control of outdoor lighting. Street lamps outfitted with Smart-
Light devices are connected through a power line and wireless links to the
Customer Information Center. The Center allows Nesa managers and users
c04
126
October 18, 2007
ENTERPRISE APPLICATIONS
16:57
Char Count=
Figure 4.10 Nesa’s Corporate IT Application Configuration. Source: Engsbro and Westmose, 2005
Figure 4.11 Nesa’s IT Processes for SmartRead System. Source: Engsbro and
Westmose, 2005
to control over a half million street lamp units. SmartLight has the following
features:
r Field teams through real-time flows of information know the conditions

of lighting fast and accurately.
r Through remote lighting options, the customer automatically adjusts and
controls large groups of street lamps.
r Customers can easily learn about a street lamp’s repair status by phoning
or e-mailing the Info Center.
r Customers can see the street lamps’ technical details in real time including
maps of lighting functioning in geographical areas.
Advantages are improved efficiency, reduced cost, and an automatic

documentation archive, which can be provided to customers, citizens, the
government, and contractors. The role of GIS is to summarize lighting
127
Figure 4.12 Nesa’s IT Processes for Outdoor Lighting Customer Information Cen-
ter. Source: Engsbro and Westmose, 2005
performance into maps, conduct spatial analysis and modeling of changes and
trends, and give Nesa employees and customers real-time access to lighting
maps.
Nesa has the SAP CRM module and other front-end systems for marketing
and sales that are shown in Figure 4.10. GIS interacts with them to provide
mapping and spatial analysis of prospects, customers, and Nesa facilities. At
the nesa.dk, customers can download maps, such as underground cabling for
their sites, which are useful for construction projects.
In sum, Nesa taken an enterprise approach with leading software packages
including many of the enterprise applications emphasized in this chapter.
They work together in an integrated manner with GIS. GIS is called into
play for those user needs where mapping and spatial analysis are relevant and
useful. The company’s IT goals of efficient and effective customer support,
system integration, and standardization are being met. At the same time, the
REFERENCES 129
architecture allows new applications such as SmartLight to be brought into

the enterprise framework to take advantage of its synergies.
CHAPTER SUMMARY
Enterprise applications are mainstays that control the “back office” of

businesses. They support and provide control for the operational business
processes. Major enterprise applications covered in this chapter are ERP,
CRM, Supply Chain Management, and Data Warehouses. GIS is also an
enterprise system which can function in a collaborative, coupled environ-
ment with the other business enterprise applications. At the underlying
storage level of databases and data warehouses, spatial features can be
built in.
The contribution of spatial technologies to enterprise applications is to
refine the accuracy of performance of the applications by recognizing lo-
cation of customers, facilities, assets, transport vehicles, and other business
phenomena. GIS also provides visualization and exploration benefits to un-
derstand enterprise information and make better decisions. Three chapter
case studies have illustrated how spatial technologies can yield these ben-
efits, and the ways that GIS is connected to work together with other en-
terprise applications. The challenge is to design spatially-enabled enterprise
architectures that provide added value to corporate users and customers, and
are flexible enough to change with the rapid technology advances in this
field.
REFERENCES
Chilsholm, Patrick. 2005. “RFID.” Military Medical Technology. Online Edition.

http://www.military-medicial-technology.com/print article.cfm?DocID=431.
Codd, E.F. 1995. “Twelve Rules for On-Line Analytic Processing.” Computerworld,
April 13.
Dyché, Jill. 2002. The CRM Handbook. Boston: Addison-Wesley.
Engsbro, Lars, and Poul Westmose. 2005. “The Changing Role of IT.” Paper pre-
sented at Sapphire 2005, Copenhagen, Denmark, April 26.
ESRI Inc. 2006. “How ESRI Technology Interfaces with SAP’s ERP Software.”
ArcNews, Winter 2002/2003. Available at www.esri.com.
ESRI Inc. 2006. NESA Deploys a Complete Business Solution: Denmark’s Largest
Electricity Company Implements Enterprise GIS. Redlands, CA: ESRI Inc. Avail-
able 8/11/06 at www.esri.com.
ESRI Inc. 2006. ESRI and SAP. Redlands, CA: ESRI Inc. Available 8/11/06 at
www.esri.com.
ESRI Inc. 2006. Spatial Data Warehousing. Redlands, CA: ESRI Inc. Available
8/11/06 at www.esri.com.
ESRI Inc. 2006. “American Suzuki Motors Corporation: Internet Mapping Helps
Customers Find Stores.” Case Study available at www.esri.com.
Farley, Jim. 2004. “Oracle 10g: A Location-enabled Platform for Enterprise GIS and
Core Business Applications.” Earth Observation Magazine, April. Available at
www.ecomoline.com.
Financial Services Technology. 2006. “Don’t Box Me In. Interview with Janice P.
Anderson of Onyx Software.” Financial Services Technology, No. 2.
Fjermestad, Jerry, and Nicholas C. Romano Jr. (eds.). 2006. Electronic Customer
Relationship Management. Armonk, NY: M.E. Sharpe Inc.
Francica, Joe. 2005. “BC Hydro Integrates ERP Solutions with GIS.” Directions
Magazine, March 19. Available at www.directionsmag.com.
Galuszka, Peter. 2005. “Lessons from Baghdad.” Chief Executive, Vol. 208, May,
p. 48.
Greenberg, Paul. 2004. CRM at the Speed of Light, 3rd Edition. New York: McGraw-
Hill.
Houwen, Peter, Wayne Hewitt, Ben Ralph, and Colin ‘t Hart. 1999. “The Keys
of Successful SAP/ESRI Interfacing.” ArcUser, October–December. Available at
www.esri.com.
Inmon, W.H. 1992. Building the Data Warehouse. New York: John Wiley and Sons.
Jukic, Nenad. 2006. “Modeling Strategies and Alternatives for Data Warehousing
Projects.” Communications of the ACM, 49(4): 83–88.
Keating, Gordon N., Paul M. Rich, and Marc S. Witkowski. 2003. “Challenges for
Enterprise GIS.” URISA Journal, 15(2): 23–36.
Kramer, Gail, and Stuart Nisbet. 2006. “The Practical Union of OLAP Analy-
sis and Geographic Mapping.” ArcUser, January/March. Available at www.esri.
com.
Laudon, Kenneth C., and Jane P. Laudon. 2006. Essentials of Management Informa-
tion Systems. Upper Saddle River, NJ: Prentice Hall.
Marakas, George. 2002. Decision Support Systems, 2nd Edition. Upper Saddle River,
NJ: Prentice Hall.
McKinney, James. 1999. “GIS Solutions for SAP R/3.” Paper presented at Sapphire
1999, Nice, France.
Mohraz, Karim. 2000. “Geographical Analysis in SAP Business Information Ware-
house.” Proceedings of the 8th ACM International Symposium on Advances in
Geographic Information Systems, Washington, D.C.: Association for Computing
Machinery, pp. 191–193.
Moore, John. 2005. “RFID’s Positive Identification.” Federal Computer Week, 19(11):
53, April 18.
Morales, Diane K., and Steve Geary. 2003. “Speed Kills: Supply Chain Lessons from
the War in Iraq.” Harvard Business Review, November, pp. 16–18.
REFERENCES 131
Nesa Inc. 2006. 2005 Annual Report for Nesa. Gentofte, Denmark: Nesa Inc.
Nogueria Barrionuevo, Rodrigo. 2002. “A SAP and GIS Integration.” Imagem Sen-
soriamento Remoto, June 21.
Oracle Corporation. Oracle Spatial Database 10g. Available at www.oracle.com.
Osterland, Andrew. 2000. “Blaming ERP.” CFO.com Magazine. Available at
www.cfo.com.
Reynaert, Bart, Patrick Vercruyssen, and Rene Horemans. 2004. “GIS and Beyond:
Looking to SAP through the Geo-Window.” Paper presented at ESRI International
User Conference, San Diego, August, 18 pp.
Reynaert, Bart. 2005. “GIS and Beyond: Integration with SAP Improves Business
Processes.” ArcUser, 8(3): 20–23.
Rittman, Mark. 2006a. “GIS-Enabling Your Oracle Data Warehouse.” DBAzine.com,
April 18. Available at www.dbazine.com.
Rittman, Mark. 2006b. “An Introduction to Oracle Warehouse Builder 10g.”
DBAzine.com, August 19. Available at www.dbazine.com.
Ross, Jeanne W., Peter Weill, and David Robertson. 2006. Enterprise Architecture
as Strategy: Creating a Foundation for Business Execution. Cambridge, MA:
Harvard Business School Press.
Roussel-Dupré, Stephanie. 2002. “What are your Customers Telling You?” Integrated
Solutions for Retailers, March.
SAP. Salzburg AG. “SAP Customer Success Story: Utilities.” Available at
www.sap.com.
SAP. Paz Oil Company Ltd. 2006. “SAP Customer Success Story: Oil and Gas.”
Available at www.sap.com.
Schnur, Mark T. “Mapping Data in SAP-Business Warehouse.” Paper available
8/11/06 at www.esri.com.
Sonnen, David, and Henry Morris. 2005. “ESRI: Extending GIS to Enterprise Ap-
plications.” White Paper, February, Framingham, MA: International Data Corpo-
ration.
Sun Microsystems. 2005. “RFID Keeps U.S. Troops Well Stocked.” Available
5/12/2005 at http://www.sun.com/solutions/documents/articles/go rfid dod.xml.
Swoyer, Stephen. 2005. “RFID Keeps Customer Service Stellar, ROI Strong. BI case
Study.” Business Intelligence Journal, Fall, pp. 35–38.
Tan, Pang-Ning, Michael Steinbach, and Vipin Kumar. 2005. Introduction to Data
Mining. Boston: Addison-Wesley.
Thurston, Jeff. “GIS/IT Integration Maximizes Enterprisewide Benefits.” GeoWorld,
16(7): 34–37.
Van Berkel, Jan. 1997. “Data Warehouse—Where to Locate GIS.” Proceedings of
1997 ESRI User Conference. Available 7/2005 at gis.esri.com/library.
Wachter, Susan, Michelle M. Thompson, and Kevin C. Gillen. 2005. “Geospatial
Analysis for Real Estate Valuable Models.” In Pick, James B. (ed.), Geographic
Information Systems in Business, Hershey, PA: Idea Group Publishing, pp. 278–
300.
Wallace, Butch, and Eric Fulcher. 2004. “Bridging the Gap Between GIS and ERP
at Alagasco.” Paper 1642. Proceedings of 2004 ESRI User Conference. Redlands,
CA: ESRI Inc. Available at www.esri.com.
Williams, David. 2004. “The Strategic Implications of Wal-Mart’s RFID Mandate.”
Directions Magazine, July 29. Available at www.directionsmag.com.
Winslow, Jon, and Anthony Lea. 2002. “Customer Relationship Management: Loca-
tion Maximizes Return on Investment.” GeoWorld, 15(4): 32–34.
CHAPTER 5
CUSTOMER-FACING GIS: WEB,

E-COMMERCE, AND MOBILE
SOLUTIONS
INTRODUCTION
Today’s consumers are increasingly dependent on web and mobile technolo-

gies. The spatial aspects of this trend are the focus of this chapter. Compari-
son is made of traditional client-server architecture with that of spatial web
services and mobile technologies. Impacts of the new architectures on com-
panies, users, customers, and the general public are considered. The chapter
shows the benefits of these more accessible and flexible approaches through
discussion, examples, and case studies.
After a comparison of the architectures, spatial web services are examined
from a business and strategic standpoint. Contrasting uses are evident in the
examples of Edens & Avant and the Delaware Department of Transportation.
Mobile services offer dynamic handheld spatial content for employees or
consumers anywhere, anytime—working, in the field, traveling, or at leisure.
DS Waters’ field workers sell water products depending on spatial for location
and routing, while Seaspan International, a Canadian tugboat/barge firm, uses
mobile mapping systems for navigation, dispatch, and fleet management.
The MotionBased case demonstrates how combined web and mobile
services can expand into a new market niche, in particular monitoring and
mapping of training results of high-performance athletes. The section on

133
134 CUSTOMER-FACING GIS: WEB, E-COMMERCE, AND MOBILE SOLUTIONS
consumer web services gives the history and impacts of the expansion of
mapping on the web, that includes mashups and open source applications.
The spatial web service and mobile architectures impact electronic com-
merce and mobile commerce. This has occurred more for business-to-
customer (B-to-C) than in business-to-business (B-to-B) e-commerce. Ex-
amples include MasterCard and Zillow.com. The trends influence markets
through gains in efficiency. Everyday examples of the spatially-enabled
smart trolley and pedestrian mobile navigation demonstrate micro-market
efficiencies.
Movement from client-server towards the web services and mobile envi-
ronments depends on having standard technical definitions and concepts. As
with the advent of other technologies, such as PCs, cell phones, and video
conferencing, standards for the spatial web take time and effort of standards
bodies to gain critical mass and acceptance. This is especially necessary when
the technology involves global sharing of information in real time. The spatial
standards and protocols from GIS vendors, the OGC, ISO, and other leading
standards organizations set a good basis for standardization and are increas-
ingly adhered to. The chapter talks about the role of several of the leading
spatial web standards, although full coverage of multiple standards is beyond
the book’s scope.
ARCHITECTURAL APPROACHES TO SPATIAL TECHNOLOGIES
This section examines and compares the major architectures for spatial appli-
cations of client-server, web services, and mobile services (sometimes called
location-based services). This does not exhaust the alternative architectures,
but they are the most common. Standalone architecture was already intro-
duced in Chapter 2. Grid computing, a promising architecture for the future,
is examined later in the chapter.
Client-Server
Client-server architecture in its simple concept, as shown in (a) in Figure 5.1,
consists of software resident on a server that accesses data in a database or
data warehouse. The server, or servers, communicates information across the
internet to multiple clients. The client may have little or no software and data
of its own (referred to as thin client) or may have moderate to high amounts
of software and data (called a thick client). The clients vary in arrangement
with the server for processing and data. Some clients are self-supported;
some process software applications in collaboration with server software;
and others depend on processing by server software (Dodd, 2005).
Take an illustration of ABC Company that follows the client-server concept
in Figure 5.1(a). The firm might have specialized servers, one for networking,
one for applications software, and one for data. Its fifty clients vary, with
ARCHITECTURAL APPROACHES TO SPATIAL TECHNOLOGIES 135
Figure 5.1 Client-Server, Spatial Web Services, and Mobile Services Architec-
tures. Source: Modified from Lopez, 2004
its engineers utilizing thick clients, while its clerks and customer-service
people use thin ones. Client-server’s strengths are: (1) it concentrates data and
software in the server(s) rather than duplicating it on client computers, (2)
it scales from a small single-server arrangement with few clients to multiple
servers and clients, and (3) it still provides local functionality if they need it.
The network can consist of a variety of types, protocols, and physical media,
which are not discussed here (Dodd, 2005). However, the movement is from
older forms of wide-area-network protocols towards the internet’s TCP/IP
protocols.
Internet for internal users in an organization is referred to as the intranet.
It follows the client-server model and is shown in (b) in Figure 5.1. In this
case, the server or servers and the clients are restricted to one organization. In
the illustration, they are all from ABC Company. This type of network offers
greater security and the ability to utilize proprietary software and data with
lower risks of loss, piracy, and intrusion (Dhillon, 2007). Spatial applications
and data can reside on ABCs servers or on client systems. A disadvantage for
spatial applications is that the data from the vast public sources that reside
on government or third party sites would have to be imported and stored on
the intranet, rather than being directly accessible. For some organizations, the
intranet offers a secure spatial architecture that appeals such as in defense,
the military, and intelligence, and to businesses that emphasize R&D on
proprietary products.
Many firms get around this by allowing its users access to both the inter-
net and intranet. Another alternative is to offer the extranet, a variation on
the intranet that allows approved outside users to have access. Client-server
for the internet is seen in Figure 5.1(c). This is similar to the intranet, but
opens up the TCP/IP-based network fully to publicly accessible servers. This
model has the advantages of vast access to information and processing, while
being exposed to enlarged security and privacy risks and greater potential for
maintenance problems or failure. This model can support web services for
users and customers, as well as other types of transactions and information
exchange.
The spatial web services architecture, seen in (d) in Figure 5.1, is a subset
of client-server internet in (c). An organization has a specialized server for
spatial web services, and other application servers. The internet is open to
the world as in (c), but now it particularly includes access to a variety of
spatial web servers. The potential for collaborations, processing sequences,
and exchanges of the organization’s spatial web servers and these multiple
public spatial servers and other web servers is the underlying driver of this
approach.
The last variant shown in (e) in Figure 5.1 is Mobile Services (Location-
Based Services). It is an open internet environment of the host organization
and other public spatial web servers. In this case a gateway connects one or
more of the host’s servers through a wireless network to mobile clients (Lopez,
2004). This architecture has the pluses of spatial web services from (d), and
the added advantage of access to spatial web services for mobile users who
move around dynamically in space. The wireless network opens up greater
security problems, as radio transmission of wireless is more vulnerable to
intrusion than other transmission modes (Dodd, 2005). Another disadvantage
is the ergonomic problem of spatial displays on handheld mobile devices.
TABLE 5.1 Comparison of Web Services and Client-Server Models
Web Services Model Client-Server Model
User Management
User profiles X X
Security and access control X X
Clients
PCs (standard and browser-base apps) X X
Mobile devices (cell phones, PDAs, etc.) X X
Middleware
API-based middleware interactions X X
Interactions with multiple servers X X
Servers
Web servers X X
Proprietary servers NA X
X = Fully supported
X = Partially supported, with constraints
Note: API is Application-programming interface.

Source: Ismail et al., 2005.
Maps can be seen but complex ones at larger scales can only be viewed by
moving around smaller window portions of the map.
What then are the differences between the client-server and web ser-
vices models? As summarized in Table 5.1 (Ismail et al., 2005), the client-
server model has robust physical security and access controls, partly due its
earlier advent allowing more time for the security procedures and standards
to become accepted. Since the web services environment is more open to the
public, potential intrusion and threats are greater, in contrast to a more re-
stricted client-server model, that reaches its extreme in a fully closed intranet
(Dhillon, 2007). Another difference is how users are added. For client-server,
user set-up is done in a proprietary way to grant a client user access to the
host’s server(s). In the illustration of ABC Company, the firm arranges for a
client Smith to be set up to begin use. However, for web services, Smith’s
user profile resides on a catalog server, which enables her quickly and auto-
matically to be connected to other web services (Ismael et al., 2005). Another
difference is that mobile devices are better supported in the web services
environment. The traditional client-server doesn’t include radio transmission
nor special servers suitable for mobile devices.
The client-server model often has multiple servers that work together
to provide information, so the model needs middleware to coordinate the
host servers. Middleware is software that connects software applications or
components, such as between servers or between the client and server. The
web services model also has moved in that direction as services become
more mainstream and enterprise-based. On the other hand, the openness of
the web services approach is based partly on availability of software code to
developers through APIs. An Application Programming Interface (API) is a

“set of interfaces, methods, protocols, and tools that application developers
use to build or customize a software program” (Wade and Sommer, 2006).
For client-server the code is more likely to be proprietary (Ismail et al., 2005).
Finally, proprietary software and data servers are more prevalent for client-
server, in contrast to the openness of software and data for web services. In
short, the traditional client-server approach is more secure and proprietary,
while its variant, web services, is open, flexible for access, and less secure.
The two models have similar underlying technology components, but they
are set up, organized, controlled, and restricted differently.
Web Services
This section delves deeper into web services. It discusses more about the
way web services functions, its business advantages, and gives two examples.
Three key elements in understanding web services functioning are the user
profiles stored on a catalog server, the ways to access web services, and the
orchestration of web services. A user profile is the characteristics needed to
set him/her up for access to a server. This includes the user preferred login
IDs and passwords, authorizations to access websites, user preferences such
as map styles, appearance of imagery, calendar, and characteristics of the
user’s mobile devices (Ismail et al., 2002). For traditional client-server, the
computer service staff gathers the user profile and sets it up. For web services,
the user profile is stored in a standard format as the Universal User Profile
(UUP). Numerous UUPs are stored on special catalog servers, many located
at service providers such as Microsoft. NET Passport or AOL Magic Carpet
(Ismail et al., 2002).
The user’s perspective is seen in Figure 5.2 which doesn’t show the full web
services architecture but only the user’s view of web services. The Catalog
Server has numerous UUPs including the user’s (face symbol shown). As a
complex chain of web service procedures take place, each web service in
turn checks with the catalog server, locates the user’s UUP, and obtains the
permissions and user profile for that service to proceed. Since this is done
automatically without the user’s human intervention, multiple service tasks
can be done very rapidly.
The orchestration of web services refers to the series of access steps that
comprise complex web services (Ismail et al., 2002; Zhao et al., 2007). The
processes are governed by protocols, mostly ones from the Open Geospatial
Consortium (OGC). Some of the important protocols for web services are
discussed later in the Standards section.
The second key function is for the user to find a desired web service. This
can be done in several alternative ways (Zhao et al., 2007): (1) by the user
searching a registry of web services located on a special web server, (2) by
an indexer, a software agent that searches systematically through websites for
services, or (3) by peer-to-peer searching, in which case there does not have
Insurance
Catalog Ratings EPA USGS NASA
Server from ABC Website for Website for Website
(holds Global Environ. Landform For Weather
UUPs) Insurance Maps Maps Maps
Third Party
Aggregator of
USGS, NASA Data
User's PDA and Workstation

Figure 5.2 User’s Perspective on Web Services. Source: Modified from Ismail
et al., 2004
to be a centralized registry or an indexer, but rather the query information

is passed along through each node in the network, eventually returning hits
back to the requester.
The last key function is orchestration, which refers to the series of access
steps that comprise complex web services (Ismail et al., 2002; Zhao et al.,
2007). The processes are governed by protocols, mostly ones from the Open
Geospatial Consortium (OGC). Some of the important protocols for web
services are discussed later in the Standards section.
This is done through service chaining, sometimes called orchestration.
An example of a sequence of web service calls is seen in Figure 5.3 that
determines land susceptibility to landslide in a risk model (Zhao et al., 2007).
The orchestration can involve more complex processes involving multiple
pathways, and steps that loop back based on logic (see Zhao et al., 2007). The
user may or may not be involved in directing this orchestration. The benefit
Figure 5.3 Orchestration of Web Services for Model of Land Susceptibility. Copy-
right 2007, IGI Global, www.igi-pub.com, reprinted by permission of the publisher
of complex orchestration is that, analogous to a macro in a spreadsheet,

many services can be automatically accessed, information processed, and
data accessed, leading to a complex end response based on a sequence of
heterogonous services, all without the end user being aware.
From a business standpoint, spatial web services unlocks the potential of
legacy systems, offers new approaches to problems and can uncover revenue
sources. One of the well-known advantages of web services for IT has been
its ability to “unlock” the information in legacy-based systems. Large-scale
legacy systems are older enterprise applications that run on large servers
programmed in an inflexible way, so it is costly to enhance or rewrite them.
The problem is much greater when several legacy systems need to work
together.
Web services can be added as a layer of software located on top of such a
legacy system(s) that connects the legacy’s inputs and outputs to web service
applications. It may not be feasible to realize on the web the full functionality
of the legacy system by this approach, but if the most important legacy
processes can be opened up to web services, that greatly expands the access
to the legacy data and the flexibility of use. Upgrading of the web services
can be accomplished relatively quickly and inexpensively.
Although large-scale, traditional GIS software is newer in general that
its IT legacy counterparts, the same philosophy can apply. A web service
for a proprietary, traditional GIS, in certain cases, can “open up the silo to
enterprise IT architecture” (Uleman, 2006).
If a company has converted its IT to a web services architecture, it is even
easier to spatially-enable it. New programming language standards especially
Geographic Markup Language (GML) from the Open Geospatial Consortium
are designed for the spatial enablement of web services. Consider the follow-
ing example from Uleman (2006). An set of steps to process a home utility
repair were implemented as a web service as follows.
r Validate the service agreement

r Assign a repair person
r Contact the homeowner
r Examine the home without service
r Repair the problem
Spatial processing steps can be written in GML to modify the process as

follows (the GML steps are in italics):
r Validate the service agreement
r Assign a repair person—based on physical proximity
r Contact the homeowner
r Plan the repair person’s routes based on the assignment
r Examine the home without service
r Repair the problem
r Monitor the repairs in the neighborhood and city, and detect unusual
spatio-temporal clusters of repair problems
Here, by adding GML code, spatial-enablement can occur quickly and

flexibly.
What are the pluses and minuses of the spatial web services approach?
Among the pluses are: speed, low cost, flexibility, wide accessibility, and user
friendliness. The minuses include the following: (a) Limited spatial function-
ality. The features currently available for spatial web services are more limited
than for traditional GIS software. For some APIs such as Google Map, they are
highly limited. (b) Size of spatial data. Some high-resolution satellite imagery
and maps with thousands of polygons may be cumbersome to transport on the
internet, even with broadband. However, this can sometimes be overcome by
compressing the format of the imagery or map. (c) Because spatial analysis
functions are not yet standardized in the real world, spatial analysis involving
multiple web services in succession may not be consistent. This problem can
be gradually resolved by more widespread acceptance of standards.
This example demonstrates the flexibility and ease of adding spatial web
services. Yet it is often difficult or an organization to change its IT and net-
working traditions. How does an organization proceed to adopt spatial web
services? Hagel (2007) strongly recommends that an organization do this
as a staged development approach. Systems development, the focus of the
next chapter, involves a series of stages leading to implementation. Hagel
recommends that a company develop a small spatial web service application
initially, gauge user responses, make improvements, and gain institutional
confidence, before advancing to broader and more complicated applications.
Dell Computer, which is often touted as successful in non-spatial web ser-
vices, followed this approach in implementing web services for its inventory
management systems (Hagel, 2007). It started with a small project, and pro-
ceeded in gradually more complex stages. Dell further convinced its suppliers
to simultaneously follow identical stages.
Hagel cautions against trying to reach too far and too fast with spatial web
services. Rather there needs to be a strategy of implementing in practical
short-term steps, while keeping in mind a full long-term solution (Hagel,
2007).
An example of spatial web services is Edens & Avant (E&A), a medium-

sized shopping center developer that serves nineteen Eastern states and over
170 shopping centers, with assets of over $3 billion. E&A built a spatial
web service for its retail customers who seek shopping-center space that al-
lows them to analyze and compare shopping centers in order to choose a
location. Online, the customer can see the layout of the center, proximity
to competitors, road maps, shortest routes, drive times, local geography, de-
mographics, and the market size. The application utilizes ESRI’s Business
Analyst, RouteMAP, the ArcIMS spatial web server (Beitz, 2001). An E&A
customer can look through hundreds of possible locations and weigh char-
acteristics of the shopping center, demographics, distances, appearance, and
layout (see example in Figure 5.4). E&A cites the benefits to retailers of
saving time in siting decisions and being able to coordinate multiple sites.
Another benefit is for E&A leasing representatives, who now have to think
proactively about the same spatial choices that confront their customers. The
E&A example shows how web services can combine a lot of data from the
Business Analyst and other sources with powerful routing algorithms, and
yet be very user-friendly and accessible to users.
A second example is a spatial web services for SDSS at the Delaware
Department of Transportation. It manages the movement of oversize and
overweight vehicles throughout the state (Ray, 2005). The design seen in
Figure 5.5 depends on extensive routing, street, bridge, and overpass height
data stored in a spatial database. The three programming modules are written
in XML (a web-based function for transporting data) and J2EE Servlet (a
web-compatible version of the Java language). The location management
module analyzes characteristics of vehicles at particular locations, while route
management considers possible routes that are feasible for these huge vehicles
in getting to a destination. In the core module, the algorithm for control and
route evaluation makes the optimal decision on the best route. The information
is output through a system interface to a spatial web server that provides
maps and data to users throughout the DOT and in transport companies.
The system has been successful and has led to lower processing costs to
estimate routing, reduced permitting time, and safety benefits (Ray, 2005).
The system was developed within the rubric of existing IT infrastructure. It
runs alongside other web services applications and can be easily modified.
Delaware DOT illustrates that spatial web services can fill new organizational
niches inexpensively yet have considerable analysis and modeling capability.
CONSUMER WEB SERVICES
The introduction of Google Map in February of 2005 had major impact not
only on spatial map services but on the entire geospatial industry. Although
not the first such service, Google Map’s appeal, ease, world coverage, and
CONSUMER WEB SERVICES 143
Figure 5.4 Edens & Avant Spatial Web Services: Property in Port Plaza Shopping
Center, New York. Source: Edens & Avant, 2007
free cost moved it to immediate prominence and opened a new geospatial era
for public consumers. Google Map can display nearly the entire earth with
satellite imagery. For urban areas it has limited features such as place names,
parks, rivers, landmarks, streets, and hospitals. Users can toggle around the
map and zoom in and out. They can get directions to addresses. A powerful
feature is the business search, that builds on Google’s strength in search
Figure 5.5 Logical Components of Routing System for Oversize, Overweight Ve-
hicles in Delaware Department of Transportation. This article is available electron-
ically as in press, August 15, 2005, from Decision Support Systems, Julian Ray,
Copyright Elsevier 2005.
technologies. A search for say, accountants in Canton, Ohio, will show with
map pushpins the locations of accountants, and web links to information about
them. Businesses can create free listings so users can pinpoint them, and they
can create “free coupons,” a small pop-up printable advertisement. What was
surprising was the success of this new “lightweight” product with everyday
consumers. The examples given in the last section were user-friendly web
services for consumers, but ones with considerable analytic and mapping
capability designed by professional staff and consultants. Google Maps and
other consumer map services are different in their ease of development. That
was made possible by the company’s simultaneous release of the free API
(Application Programming Interface) for Google Map. The API enabled any
user to modify code and customize display not only the standard Google map
layer but additional map features, a phenomenon known as mashups. The
consumer can superimpose data from one or more web sources on standard
imagery (Roush, 2005). The term “mashup” comes from hip-hop music, in
which digital equipment can blend one song’s vocal line with background
music from another tune (Ramsey, 2006).
This success of consumer spatial web can be explained by the theory
in information systems of the Technology Acceptance Model (TAM). This
theory attributes the acceptance of a technology to two main components:
“perceived ease of use” and “perceived usefulness” (Davis, 1989, 1993).
The consumer offerings of the spatial web have both components. It can be
argued that desktop GIS is more complex and not so user friendly, hence the
widespread adoption of the web services.
Microsoft, which before Google had offered less popular map services
called Map-Point Location Server and Microsoft Streets and Trips, followed
CONSUMER WEB SERVICES 145
suit in 2005 by introducing Microsoft Virtual Earth, and Yahoo introduced

Yahoo Maps the same year, both with the APIs available for free. Google
shortly added a slightly more sophisticated mapping service, Google Earth,
which provides a 3-D view of the earth and additional basic mapping capa-
bilities.
Since the APIs were freely available, a vast number of developers as well
as everyday amateurs have stepped in and developed customized versions
of these services, re-marketing them. Tens of thousands of mash-ups have
appeared, some of which have gained prominence and considerable visitor
base. The explosion in these offerings opened up a new niche of spatial
consumer users and ignited the interest in mapping across broad reaches of
the internet public.
Consumer mapping has weaknesses and drawbacks. One is reduced spatial
functionality. For instance multilayer displays have been limited and such
software functions as coordinate systems and projections are not present.
However, this is somewhat improved by third-party firms such as PlaceBase
that provide more functionality (Vermes, 2006). Mainstream GIS software
vendors have felt a wake-up call from this consumer mapping explosion. ESRI
has introduced ArcWebExplorer which competes in the free, low-endmarket
space, but with more spatial features and even some analysis. This is a very
dynamic business area that is only at the start of its product curve. This
section now turns to an example of a low-end spatial consumer services and
a somewhat more sophisticated, free web service from a non-profit.
Trulia.com
Trulia.com is a residential real estate search engine that gives free real es-
tate pricing and other information in the current real estate market. Maps
and graphs display information including average home prices by size of
home, market price trends, community demographies, neighborhood popu-
larity, comparable listings, and much more. One feature is a map display that
allows the user to see all the listed properties for sale. When visitors click on
a property pushpin, the listing price appears alongside corresponding real es-
tate descriptions (see Figure 5.6). “Heat maps” are Google mashups that show
the levels of current pricing for a larger city or county. Trulia’s friendly and
well-designed user interface makes the best of the functionality constraints
of the Google Earth API.
DataPlace.org
A second example is DataPlace (www.dataplace.org), an initiative of Knowl-
edgePlex, that is sponsored by the Fannie Mae Foundation, a nonprofit in-
terested in providing mortgage and housing information to the public. It is a
spatial web service that provides housing and demographic information and
maps across the United States by city, community, and census tract. The data
sources are the U.S. Censuses of 1990 and 2000, federal home mortgage
Figure 5.6 Trulia Real Estate Search. Source: Trulia, 2007
data, and special federal tabulations on housing needs by income level of

households. Users can create extensive city, neighborhood, and area profiles
by housing, demographic, income/employment, and mortgage lending data.
They can be mapped thematically, as seen for year 2000 housing vacancies
in Evanston, Illinois (Figure 5.7). Using the hundreds of variables available,
a user can accumulate a spatial tapestry of a neighborhood or community.
This application was created with the support of a nonprofit entity and ini-
tially intended for an audience of affordable housing community specialists.
MOBILE SERVICES 147
Figure 5.7 DataPlace.org Map of Vacancy Rate in Evanston, Illinois. Source: Dat-
aPlaceTM. 2007. World Wide Web (http://www.dataplace.org/map/?z=1), February
26, 2007
The mapping software came from PlaceBase Inc. that has deep expertise
in geospatial web platforms and databases. In contrast to Trulia, DataPlace
emphasizes nonprofit-private collaboration.
MOBILE SERVICES
Mobile services (also referred to as Location-Based Services or LBS) are

a wide variety of services that provide access to locational and associated
attribute information for users of portable locational devices. Spatial mobile
devices communicate with server networks by GPS, RFID, Wi-Fi, and WiMax
(Astroth, 2005). A device can have an active or passive GPS unit built in,
allowing it to determine within feet its X,Y coordinate location. The X,Y
coordinate location can be transmitted to a mobile server through Wi-Fi or
WiMax communications. A device with an RFID tag can have its location
read by a nearby RFID reader and the reader updates information on a server
with the reader’s X,Y location.
Mobile services benefit a range of businesses by providing a mobile em-
ployee with maps, web-based information, knowing where he/she is located,
communicating instantaneously, and receiving field data inputs, resulting in
enhanced responsiveness to customers in the field. While computer users are
more fixed in location, the users of mobile services are in dynamic physical
movement. Some business examples of spatial uses are employees involved
in monitoring assets in the field, logistics, navigation, real estate, marketing,
sales force, and electronic marketplaces.
A common model for mobile services is shown in Figure 5.1e. A web
service network is connected via a gateway to a wireless network, which
communicates with mobile devices. If the mobile device has GPS or other
georeferencing, it can transmit locational data for mapping back to the server.
In some situations, data are transmitted manually between the field and server
network. There are multiple alternative mobile networks that are beyond the
scope of the book (Dodd, 2005). Several examples of them are:
r A network of portable RFID readers that transmits inventory information
through a wireless network to a gateway and then to a client-server
network. An application server analyzes the inventory characteristics
and records the location of the inventory for spatial analysis.
r OnStar or NavStar devices in cars, which communicate via a wireless
network to a client-server network connected to a spatial server.
r Enhanced 911 (E-911) network. In an emergency, a cell phone which is
GPS-enabled sends the coordinate location of the phone through a cellular
network to a 911 Public Safety Answering Point (PSAP), which routes
the call to the closest location to the caller for dispatch of emergency
services.
In mobile networks relevant to this book, the messaging needs to include
spatial information. This might be the coordinate location of the mobile
device that is detected by GPS, RFID, and sensors. Spatial information in the
form of maps and spatial analysis outputs can flow to and from the mobile
device. Non-spatial information usually accompanies the spatial information.
For instance, a field utility worker may transmit back to the server network
the X-Y coordinates of utility assets that need maintenance, along with a text
description of the repairs required. Other types of common messaging for
mobile systems include address verification and matching, driving directions,
voice messages, and proximities.
Some of the qualities desired for a mobile device network (Lopez, 2004)
are:
r Mobile. The mobile device can connect to the mobile network anywhere,
anytime.
MOBILE SERVICES 149
r Real-time. The mobile devices transmit information instantaneously.

r Scalable. The mobile network can grow in size considerably without
exceeding its technical limitations.
r Sufficient performance. The network can handle a sufficient number of
mobile messages to suit the business goal.
r Interoperable. The information from mobile devices can be reliably trans-
mitted across gateways and enter a client-server or other network.
r Open. The standards and protocols are widely known, such as HTTP,
XML, Wireless Markup Language (WML), or Multimedia Markup Lan-
guage. The common standards allow the different parts of a mobile
network to work together well.
r Secure. Security features protect the network from intrusion, information
theft, or shutdown.
In designing a spatial mobile network, these qualities need to be prioritized

for the business problem and weighed against the costs.
An emerging development in spatial mobile services is location-awareness.
This refers to automatic reporting to a customer of information related to
his/her current location. For instance a customer in a downtown city neigh-
borhood can request hotel information from his/her GPS-enabled mobile
device and it will signal what hotels are within a quarter mile and also have
rooms available.
Some vendors of mobile services are listed in Table 5.2. They provide
mobile infrastructure equipment and software, GPS and other positioning
features, software for spatial analysis and mapping of mobile data, services
and content to mobile devices in the field, and web portals for business and
individual customers (Lopez, 2004). The practical development and operation
of spatial mobile networks involves working with combinations of these
vendor categories.
This section on mobile services now turns to two mini-cases of spatial
mobile services: DS Waters and Seaspan.
DS Waters
DS Waters, based in Atlanta, is one of the largest bottled water delivery
companies in the U.S. having 5,500 employees and revenues of over $1 billion.
Its brands include Crystal Geyser, Sparkletts, Belmont Springs, and Sierrra
Springs. Its employees had previously used combinations of cell phones,
radios, and pagers to communicate with dispatchers and supervisors in the
field. There were often delays of hours in responses (ESRI, 2005). Another
problem was payroll. Based on paper timesheets turned in by the field delivery
people, weekly payroll processing was time-consuming, costly, and full of
errors.
TABLE 5.2 Major Vendors by Category for Mobile Services
Infrastructure Software Providers

Provide databases, application servers, positioning servers, and enterprise applications.
IBM, Lucent, Microsoft, Motorola, Nokia, Oracle, and Sun
Network and Handset Positioning Vendors
Give positioning technology in the network structure or GPS positioning technology into
handsets and other mobile devices.
Alcatel, Cambridge Positioning Systems, Ericsson, Lucent, Motorola, Nokia, Qualcomm,
Siemens, Trimble
Specialty Tools for LBS
Offer mapping, spatial analysis, routing, and geocoding software for deploying mobile
services
ESRI, Autodesk, Ionic, Intelliwhere, MapInfo, Telcontar
Mobile Services
Provide branded custom mobile services.
Mapping, driving, and yellow pages
MapQuest, Vicinity, Webraska, Yahoo
Satellite imagery services
GlobeXplorer
Real-time traffic
SmartTraveler, TrafficMaster
Geocoding and gazetteer
Whereonearth
Content Providers
General
Governments (federal, state, local)
Geographic
TeleAtlas, Geographic Data Technologies, Navigation Technologies
Demographic and business
Acxiom, Claritas, InfoUSA, Polk
Wireless and Web Portals
Wireless carrier portals
Verizon, AT&T, British Telecom, Vodaphone
Wireless ASP portals
InfoSpace
Telematics portals
OnStar from GM
Source: Lopez, 2004.
DS Waters implemented a mobile system based on cell phones that feeds

delivery and payroll data to a gateway and into spatial web servers, similar to
the model in Figure 5.1e. An application server can produce reports and maps
for supervisors and field workers (ESRI, 2005). The cell phones are ruggedi-
zed and utilize the WorkTrak mobile software that includes web display of
delivery routes (see Figure 5.8).
MOBILE SERVICES 151
Figure 5.8 Cell Phone and Mobile Web Service which Track Deliveries at DS
Waters. Source: ESRI, 2007
After a pretest on 100 water-cooler installation people, the system was put
into use with 2,000 delivery and maintenance employees nationwide (ESRI,
2005). As the day goes along, the field person enters his/her shift, break,
and service-job information at stop points (ESRI, 2005). After transmissions
are received at the dispatcher’s console, the dispatcher can re-deploy the
technician based on maps and reports. After finishing each job task, the
field employee receives text information on his/her cell phone regarding the
specifics of the next job.
Although the system was planned to track and analyze all the field workers’
detailed productivity daily and weekly, the managers in charge found out
instead that it produced information overload (ESRI, 2005). Instead, the key
management use has turned out to be checking on exceptions for individuals
or on bottleneck situations.
DS Waters estimates the average time saved per employee is 20 minutes
daily which adds up in added productivity to $3 million per year (ESRI,
2005). The example demonstrates the net benefits of a spatial mobile model,
and also shows that pre-planned uses may need to be modified.
Seaspan International Ltd.

Seaspan International Ltd. is a part of the Washington Marine Group, one of
Canada’s largest shipping and marine companies. With forty-six tugs and sev-
eral hundred barges, Seaspan is one of the largest marine hauling companies
in the Pacific Northwest (Miller, 2005; Seaspan, 2007). The tugs and barges
are contracted by shipping firms, developers, other companies, provincial and
federal governments, and enterprises on islands. They haul ore, logs, build-
ing supplies, and mining and other equipment (Miller, 2005). The routes are
mostly along the Fraser River but cover the entire Canadian and U.S. west
coasts.
Seaspan’s tugs, such as the one in Figure 5.9, are dispatched from Van-
couver. Previously dispatchers communicated with the tugs by radio phone
and mapped their locations by moving magnetic markers on a map board. On
the boats, all the paperwork was done by hand. Washington Marine’s CEO
made the decision that the system had to be upgraded to digital dispatch. The
captains were demanding this, and it was expected to increase competitive-
ness. The new system was implemented as a prototype within ninety days,
in consultation with ESRI Canada. The architecture of the mobile system is
seen in Figure 5.10. The data head is usually able to communicate to shore
by a cell phone wireless network, but if the vessel is too far from shore, it
automatically switches to a satellite network (Miller, 2005). The spatial appli-
cation server at the dispatch center stores and provides maps, logs, traffic, and
other information. At the dispatch center, large projectors display maps and
data on dual screens. External information can be accessed through content
servers on the internet. Privileged users including managers and families of
the crews in the tug fleet also have web access.
MOBILE SERVICES 153
Figure 5.9 Seaspan Tug Hauling a Barge Offshore Canada. Photo courtesy of
nix-pix, licensed under Creative Commons Attribution-NoDerivs 2.0, source http://
www.flickr.com/photos/nix-pix/161461860/
On the tug, the captain can select an assignment by viewing information on

the data head and confirm job assignments through messaging with dispatch.
Once confirmed, the job starts immediately, but the captain can also end it in
circumstances such as bad weather, which simultaneously stops the customer
from being charged. A second phase of the project links the time records and
contract management to the billing and financial system.
Figure 5.10 Architecture of Seaspan’s Mobile Network. Source: Miller, 2005

Figure 5.11 Map Showing Ship Located in Vancouver Region and Assigned Tugs.
Photo courtesy of nix-pix, licensed under Creative Commons Attribution-NoDerivs
2.0, source http://www.flickr.com/photos/nix-pix/71735723/
The system lets dispatchers and managers view Seaspan’s tugs and barges
from the dispatch center or online. (an example appears in Figure 5.11).
Other information available include ocean currents (from an external content
server), GPS-based vessel speeds, tug voyage history, accident reporting, tug
logs, and static orthophotos (Miller, 2005). An unexpected benefit stemming
from the post-911 security environment is that all ships within 200 miles of
shore in the U.S. and Canada must have a functioning transponder. Hence the
locations of all competitor tugs and barges in this shore range are now known
and can be overlaid on Seaspan’s maps.
In this example, a visionary CEO asked for a cutting edge spatial mobile
system. In spite of some management resistance, it was successfully and
quickly implemented and has improved safety, efficiency, and competitive-
ness.
CASE STUDY: MOTIONBASED TECHNOLOGIES
MotionBased Technologies started as a Sausalito, California, based small

firm to perform performance analysis of high-endurance athletes, such as
marathoners, long-distance bikers, hikers, and competitive sailors. The firm
was purchased in 2005 by Garmin, a GPS equipment company, and continues
as a Garmin division. An athlete customer utilizes a Garmin GPS device
during a practice session, recording his/her route (X-Y coordinates), elevation
(Z coordinate), and heart rate. A special technical contribution of the company
was to cross-reference the athlete’s X-Y GPS coordinate with a national
E-COMMERCE 155
elevation database, Global Mapper. A new Garmin device used by some

runners also provides body cadence in rotations per minute.
After an athlete completes a practice route, the athlete uploads the informa-
tion to MotionBased’s website. At the site, the athlete can use the “dashboard”
to view metrics of their practice route. For instance, for a marathoner, it can
analyze running speed versus elevation incline and heart rate throughout a
practice run. The site can also simulate activities on a variety of types of
maps such as street, topo, or elevation, depending on the type of sport. The
athlete maintains a record of his/her performance history on the same route
and different routes. Also, the athlete can compare his/her performance with
that of other athletes taking the same route.
There are a range of levels of legal rights that the athlete can retain or grant
to MotionBased regarding his/her performance history. MotionBased does
not receive fees except for a small proportion of its clients who pay a small
monthly fee for extra features. The start-up company ran at a loss, while the
value to Garmin was to acquire MotionBased’s web services skills and some
proprietary software design. The web application is programmed in Java and
uses the MySQL database.
For map services, MotionBased utilizes the APIs for both Google Maps and
ESRI’s ArcWeb Services. Many of the base maps from these services come
from the base map providers TeleAtlas and NavTeq. MotionBased’s goal is
to provide performance athletes with pertinent visual and tabular information
that they can analyze outside of the training sessions. The firm also seeks to
feed back more information to the athlete during performance, so he/she can
also make comparisons during actual practice to his/her performance profile.
This case illustrates a combination of web services and mobile services.
The proprietary original design was done by the three founding employees
in a year or so, and the design remains easily extensible and flexible. It is a
good example of the advantages of spatial web and mobile services and it
took place in an environment with no legacy systems to hold things back.
E-COMMERCE
The move towards spatial web and mobile services is influencing markets and
e-commerce. This is seen in cases and examples in the chapter. For instance,
the information available to Edens & Avant’s East Coast retail customers
influences the locations of shopping centers they choose to locate in. For
MotionBased, a market for a new service has been created.
This section considers the two leading e-commerce models of business-
to-consumer (B2C) and business-to-business (B2B). It asks the question how
can the new models of spatial web and mobile services influence B2C and
B2B. It finishes by considering two examples: MasterCard’s ATM Locator
and Zillow.com.
Business–to–consumer e-commerce involves using the Internet for re-

tail selling of products and services to individual customers (Laudon and
Laudon, 2007). Some highly successful and commonplace examples of non-
spatial B2C e-commerce are airline reservations, which are increasingly done
through travel or airline web services, and book sales by Amazon.com, Bar-
nesandNoble.com, and services, which replace visits to bookstores.
GIS and spatial technologies influence B2C in the following ways:
Customers can:
r find the locations and driving directions for elements of B2C including
retail outlets, shopping centers, service outlets
r do map comparisons of locations of items being sold such as homes
r examine the influence of the natural environment on products and services
r examine locational proximities
r spatially view characteristics of places, travel routes, cities, neighbor-
hoods, points of amenities, hazards, before deciding on a service or
product
r run spatial models and forecasts, geostatistics, and SDSSs to evaluate
products and services
Businesses can:
r do the same spatial viewing and study as customers

r understand the characteristics of customers through geodemographics
and other spatial methods
r price products and services with models that include space
r study the spatial array of their own retail and sales facilities and those of
competitors in order to reach customers and serve them better
For B2C, GIS and spatial technologies introduce new dimensions of think-
ing and understanding of products, services, and markets for buyers and
sellers.
Business-to-business (B2B) e-commerce refers to markets and sales of
goods and services between businesses (Laudon and Laudon, 2007). A highly
successful B2B is the Milacron website for companies buying and selling in
the plastics processing industry (Laudon and Laudon, 2007). On its site market
exchanges occur for some of the products and services in this industry such
as buying and selling machinery, tooling, and related supplies.
EBay.com is another example of a website that constitutes a market for
buying and selling a vast range of products and services, albeit mostly for
individuals, rather than businesses. Its exchanges include B2C and B2B. It
also supports direct exchanges between two consumers, termed C2C.
E-COMMERCE 157
GIS and spatial technologies influence B2B in the following ways:
Businesses involved as a party in a B2B exchange can:
r do map comparisons of locations of items being sold such as machinery,

agricultural products, commercial properties
r examine the influence of the natural environment on products and services
r spatially view characteristics of places, such as factory neighborhoods,
dump sites, agglomeration of hospital services, fast-food labor pools
before deciding on a service or product
r run spatial models and forecasts, geostatistics, and SDSSs to evaluate
products and services
r examine locational proximities
r view regulatory and legal jurisdictions and boundaries
r study spatial differences in costing and pricing
Businesses hosting a B2B web service can:
r have user features that divide customer pools of buyers and sellers into
local, regional, or national sub-pools
r have models of routing for deliveries of goods being bought or sold, so
customers can be better informed about delivery times
r present informational data on products and services that are strongly tied
to geography such as real estate and wine regions
r analyze geography of customers, so as to advertise their B2B service to
prospective businesses in underserved regions
For B2B, the influence of spatial web services depends partly on how
inherently geographical the product or service is. Maps and spatial results
can be made available for reference purposes to all the market parties.
Two examples demonstrate how spatial web services can influence e-
commerce.
MasterCard
MasterCard, the second largest payment system in the U.S., serves 25,000
member banks worldwide. In 2003, MasterCard decided to develop a web
service to help its hundreds of millions of card customers to locate an ATM
anywhere in the world. The MasterCard Global ATM Locator service now
provides maps and descriptions of 900,000 ATMs worldwide that are in the
Cirrus ATM Network, owned by MasterCard, as well as other ATMs. Prior to
this, MasterCard had two different locator systems with limited information
that frequently differed on locations (Chen, 2005).
The web service locator application is based on Google Map and uses street
information from TeleAtlas (MasterCard, 2007). A customer can request that
it forward a text message of directions to access the ATM location to the
customer’s cell phone. If the customer’s phone has GPS, he/she doesn’t have
to enter current location. The service also indicates the physical handicap
status of sites. Overall, the result is that accuracy has improved by 50 percent
(Chen, 2005). It has hundreds of thousands of hits per day. The service not
only makes customers’ credit card use more efficient, but also helps banks and
other credit card issuers build customer ATM volume. The spatial dimension
has altered the B2C model for credit card use at ATMs, by increasing customer
and banking information and efficiencies.
Zillow.com
Zillow.com is a leading consumer real-estate site founded by Richard Baton
and Lloyd Frink, former Microsoft and Expedia executives. From its start in
2005, Zillow has become one of the leading real estate sites. Zillow.com has
52 million estimates of house prices in the U.S. and the number is growing. The
estimates, referred to as Zestimates, are seen in Figure 5.2. They vary in accu-
racy when compared to actual sales prices (O’Brien, 2007). For instance, only
65 percent of its estimates have been accurate within 10 percent of actual
sales price (O’Brien, 2007). The site provides listing prices, sales prices, Zes-
timates, and information on properties. The mapping gives overhead satellite
views with real estate plots superimposed, showing pricing. Its “heat map”
Figure 5.12 Zillow.com Z-Estimate of Property. Source: Zillow.com

R
Zesti-
TM
mate valuations
E-COMMERCE 159
Figure 5.13 Example of Zillow.com Make Me Move Page. Source: Example of

Zillow.com
R
Make Me MoveTM Price Page
feature produces a thematic map for cities showing the “hot areas” in housing
values per square foot. A new feature “Make Me Move” allows a seller to
list a price that is sufficiently above the area’s sales norm to incentivize the
seller to move. The seller can add additional facts and information (see Figure
5.13). If a seller goes ahead with a Make Me Move, a special symbol will
appear on the map so that buyers can contact that seller.
Real estate markets are known to be asymmetric, i.e. one party (the seller)
has a lot more information than the other (the buyer) (O’Brien, 2007). Zil-
low.com has a goal to even this out with large amounts of information available
for buyers as well as sellers. Zillow is a spatial web service that is changing
the e-commerce market for real estate. If this model were applied for buyers
and sellers of commercial real estate, it would represent a B2B example. The
Sperry Van Ness case from Chapter 1 comes close to this, but the difference
is that it is not widely known across the whole commercial real estate market.
Market Efficiency
One of the dimensions of spatial e-commerce seen in the examples is market
efficiency, i.e. how buyers and sellers can utilize spatial information to make
Figure 5.14 Smart Trolley: Using a Cart Computer While Shopping.
more efficient choices. The efficiency stems from geographical insights and
from better use of space (analogous to better use of time). The geographical
insights occur because of the visualization and spatial analysis from GIS, and
the locational sensing and tagging of objects by GPS, RFID, and sensors.
Two examples of marketing efficiency for mobile devices are smart trolleys
and pedestrian navigation with cell phones. Smart trolleys in supermarkets
have a computer with screen installed on the grocery trolley (see Figure
5.14). The trolley can sense grocery shelf items having RFID passive tags.
Most smart trolleys also have bar code readers so the shopper can read bar
codes of items not tagged by RFID. In either case the trolley’s computer
displays pricing and other information about the store items. Some trolley
models let a customer enter a shopping list in advance through their home
computer. Customers with a smart trolley can also organize aisle trips. A map
on-board displays the best route through the aisles (Senne, 2005), a time-saver
in mega-grocery stores!
For some smart trolleys, customers with preferred-customer cards receive
special discounts and are directed to sales areas based on their buying patterns.
For smart trolley shoppers, there is self checkout, with a lot of the items
already pre-entered during the store trip. The cost for a store to convert to
smart trolleys is over $100,000, but there are multiple benefits for the store
and customers. For the store, checkout is quicker and the employer tasks are
automated. Also, the store has another way to advertise products, and can
get better response to its preferred-customer cards (Senne, 2005). For the
customer, the benefits are the time savings and better choices to gain value in
purchases. In the example, spatial knowledge increases market efficiency for
buyers and sellers.
Analogous to NavStar and other car navigation systems, pedestrian naviga-
tion systems inform city dwellers about locational information. This approach
is based on a cell phone having GPS navigation, precise to thirty feet or less;
an electronic compass that orients the pedestrian; and mapping web services.
One example in Japan is the GeoVector service, which partners the U.S. firm
STANDARDIZATION 161
Figure 5.15 Cell Phone Used for City Navigation and Information, Tokyo. Source:
Geo Vector, 2007
GeoVector with the Japanese cellular carrier KDDI; Japanese networking

firm NEC Magnus; Mapion, a map service; and Sony Ericsson that produces
handsets. As seen in Figure 5.15, a pedestrian in Tokyo can select a type
of destination, for instance banks, and point the phone in a direction. The
list of banks located in that direction appear on the screen along with the
distances and other information, and maps can be displayed to indicate routes
to get there. Spatial features are increasing the customer’s knowledge of retail
markets in the city and improving the efficiency of the market.
STANDARDIZATION
The final section examines standards in GIS and spatial technologies. It

considers some of the key spatial-web service standards, and why they are
important.
In its early history, GIS tended to be standalone or departmental. There
was much less networking and interoperability. Spatial software standards
depended on the GIS major vendors. Today, there are many software alterna-
tives, types of spatial web service providers, mobile service firms, networking
and communications models, and sensor and other data gathering technolo-
gies. Consequently, there is growing need for spatial and GIS standards as
well as for great acceptance of the standards.
Overall, the geospatial industry is moving towards more standardization

(Daratech, 2004; Croswell, 2005). Along with this trend is movement towards
greater interoperability and openness of software and data. Interoperability
refers to spatial applications that work across different systems and orga-
nizations. Open systems refers to the ability of a user or developer to have
free or affordable access to spatial software, software code, and spatial and
attribute data. The benefits of an interoperable and open environment include:
(1) ability to operate applications on a variety of technology platforms and
databases, (2) integration of spatial and non-spatial applications, (3) collabo-
ration among spatial technologies, (4) ease in sharing and using geographic
data, (5) greater commonality in development methods and tools in the GIS
community (ESRI, 2003).
Standards are set by leading vendors and by important standards organi-
zations. ESRI Inc. has been a business leader in setting industry standards,
especially for storage of spatial data. For flat-file storage of spatial data, its
shape file (.SHP) norm has become the leading standard. ESRI went further
with a more advanced spatial storage type of geodatabase (.MDB) that has
also become an industry standard. In spatial databases, Oracle is the most
important industry standard leader for middle- and large-sized applications.
In addition to its mainstream non-spatial Oracle relational database stan-
dards, the Oracle Spatial 10g database has set standards for access to spatial
data that are influential. Other important software and database vendors such
as Intergraph, MapInfo, Microsoft, and IBM have also contributed some
industry-based standards.
The major standards-setting organizations influencing GIS and Spatial
Systems (see Table 5.3) are governmental agencies and nonprofits. Several
of them support generic standards, such as the American National Standards
Institute (ANSI), the European Committee for Standardization, and the Inter-
national Organization for Standardization (ISO). ANSI has set programming
and processing standards that have influenced the development of the com-
puter and IT industries. The ISO has not only established many significant
computer and IT standards, but has a sub-group, ISO TC 211 with the mission
to provide standards in digital geographic information (ISO, 2007).
Among the organizations setting spatial standards, the most important and
active one has been the Open Geospatial Consortium (OGC). The Federal
Geographic Data Committee has set a goal of setting standards for a national
digital spatial infrastructure for the U.S. and has made some progress towards
that but is less influential in practical standards for spatial developers and
users.
Two key standards organizations, the OGC and ISO, have been the most
influential for the spatial industry. Their standards are particularly important
for the web services and mobile services environments.
The OGC, founded in 1994, has grown from eight charter members to 336
corporations, universities, nonprofits, and governments. By 2007, it had nine-
teen standards listed that particularly related to interoperability and openness,
STANDARDIZATION 163
TABLE 5.3 Leading Organizations for Spatial Standards
Name Type Website
American National Standards Institute standards www.ansi.org

European Committee for standards www.cen.eu
Standardization
Digital Geographic Information standards www.dgiwg.org
Working Group
Federal Geographic Data Committee standards www.fgdc.gov
Global Spatial Data Infrastructure interoperability www.gsdi.org
International Hydrographic standards www.iho.shom.fr
Organization
International Organization for standards www.iso.org
Standardization
Organization for the Advancement of standards www.oasis-open.org
Structured Information Standards
Open Geospatial Consortium specifications www.opengis.org
Open Mobile Alliance specifications www.openmobilealliance.org
World Wide Web Consortium standards www.w3.org
Web Services Interoperability interoperability www.ws-i.org
Organization
Source: Modified from ESRI, 2007.
i.e. apply to the new services era. Among the most important of these are the
six standards seen in Table 5.4.
Geographic Markup Language (GML) is the best-known modeling lan-
guage applied to geographic systems on the Internet. GML statements coor-
dinate geographic features and analysis. Web Catalog Service (CAT) specifies
common web-server catalogs for user profiles are set up across the hetero-
gonous servers on the web. The Web Map Service (WMS) defines standards
of how single and overlaid map views can be arranged and function on the
web, overcoming the problems of spatial data coming from diverse sources.
The influence of the OGC standards on spatial web services is seen in the
model of the web service standards shown in Figure 5.16 (modified from
Zhao et al., 2007). It shows the spatial data at the bottom feeding up to
data access services and integrative spatial web services. GML applies to
this whole middle section of the model as the spatial language. The spatial
catalog mentioned earlier is governed by the CAT standard at the upper right.
Gazetteer serves as a geographic dictionary. Finally, the web services feed up
to the spatial clients and their applications. It is evident that OGC standards
fully encompass the web services environment. Not surprisingly its standards
have been the most important in this realm.
The ISO’s TC211 standards group has been influential in setting the ISO
19115 standard for metadata. Metadata describes digital geographic datasets.
It is crucial for cataloging them, as the population of datasets continues to grow
TABLE 5.4 Important OGC Standards
OGC Standard Abbreviation Content
Geographic Markup GML The Geography Markup Language (GML) is an XML

Language encoding for the transport and storage of geographic
information, including both the geometry and
properties of geographic features. GML is also an
ISO standard.
Simple Feature SFS This standard specifies both a common architecture
Access Parts 1 and 2 (Part 1) and SQL schema (Part 2) that supports
storage, retrieval, query and update of geospatial
features with simple geometry via the SQL Call
Level Interface (SQL/CLI), an ISO standard. There
are also COM and CORBA binding specifications.
Simple Feature Access Part 1 and Part 2 are also
ISO standards.
Web Catalog Service CAT Defines common interfaces to discover, browse, and
query metadata about data, services, and other
potential resources.
Web Coverage WCS Supports electronic query, description, and retrieval of
Service geospatial data as “coverages” – that is, digital
geospatial information representing space-varying
phenomena, such as imagery and gridded data.
Returns the result in a standard format, such as
GeoTIFF.
Web Feature Service WFS Enables a client to retrieve and update geospatial data
from multiple Web Feature Services. The
specification defines interfaces for data access and
manipulation operations on geographic features,
using HTTP as the distributed computing platform.
The payload encoding is GML and is independent of
how the data are stored.
Web Map Service WMS Defines the operations for requesting, creating, and
displaying registered single and superimposed map
views of information input concurrently from many
heterogonous sources. Includes how to get
information on map type and content from a server.
Source: OGC, 2007.
for web services. In the grid computing environment that accesses numerous
spatial datasets globally, accurate metadata is even more a critical factor. Also
influential are ISO 19125-1 and 19125-2 for Simple Feature Access. This is
a SQL schema for digital storage, retrieval, update, and query of both spatial
and non-spatial features, and corresponds to OGC’s SFS standard.
In summary, spatial standards have become essential in the increasing
internet-based and services environment for GIS. Major vendors and standards
STANDARDIZATION 165
Browser Web Servcies Application
Geospatial Clients
CAT Gazetteer
Portrayal WICS WMC
GML Integrative Geospatial Web Services
WFS WCS WMS

Geospatial Data Access Services
Geospatial Data
Figure 5.16 OGC Standards for Spatial Web Services. Copyright 2007, IGI Global,
www.igi-pub.com, reprinted by permission of the publisher
organizations have made progress in providing a robust set of spatial stan-

dards, and they are growing in adoption.
Grid Computing
As a final topic, Grid Computing represents a future framework that offers
greater processing power and a worldwide reach of collaborative services
beyond what is present today. Grid Computing refers to “flexible, secure,
coordinated resource sharing among dynamic collections of individuals,
institutions, and resources—what is referred to as the virtual organization”
(Foster et al., 2001). It depends on open standards and protocols, that enable
access to unprecedented amounts of resources across the internet. Rather than
file-sharing, in the Grid environment, there is a collaboration of multiple web
services working together across virtual organizations worldwide (Foster,
2001).
A framework for Grid Computing, seen in Figure 5.17 (Buyya and
Venugopal, 2005) has resources and databases located globally. The users
access resource brokers, which in turn can access resources directly or link
to grid information services that provide specialized help in accessing the
resources. The information services have specialized strengths in discovering
appropriate resources. A user accesses a sequence of resources and informa-
tion services into order to solve his/her problem.
i
Grid Information Service
Grid Resource Broker
R2 Database
Application
R3 R4
R5 RN
R6
R1
Grid Information Service

i
Figure 5.17 Global Grid Computing Environment. Source: Rajkumar Buyya, Uni-
versity of Melbourne, Australia
Grid Computing is especially good at solving giant problems like dis-

aster management, financial simulation, and environmental simulation for
world regions. Because it can search everywhere for resources, it excels at
locating slack resources. Although Grid Computing is applied currently by
large governmental organizations, big science projects, and larger technical
companies, it is expected to become more widely available in the next fifteen
years. It offers GIS potential for worldwide data sharing, greater processing
power, and combinations of spatial and non-spatial web services to achieve
more sophisticated outcomes.
CHAPTER SUMMARY
The trend in GIS is towards the spatial internet bus from Chapter 1. This
includes movement towards web services, mobile services, and spatial e-
commerce. This chapter examines this growing area of spatial applications.
Starting with the architecture of the major models for GIS and spatial, it
next covers in more depth the business and commercial sides of spatial web
services.
Consumer spatial web services and mobile services are revolutionizing the
public’s awareness of spatial technologies and providing many benefits as
well as opening up new niches. The MotionBased case study illustrates the
architecture, business advantages, and practicalities of spatial web services
and mobile services. A specialized customer base can get their own data
through GPS and receive analysis of it through an interactive web service.
The environment is flexible and customer-friendly.
REFERENCES 167
The new internet-based spatial service environment is influencing e-

commerce especially B2C. It brings new dimensions to electronic markets
and creates efficiencies. The examples of MasterCard and Zillow.com demon-
strate how e-commerce markets are changed by spatial applications. Mobile
services such as smart trolleys and portable locators create new efficiencies
both for customers and businesses.
Standards are key to the success of GIS and spatial technologies. Although
they have been around from the start of the IT industry, in the web-based,
open, and interoperable environment today, they are critical. Vendors and
standards organizations have progressed in establishing a robust spatial set of
standards for the current era that are gaining increasing acceptance. The future
will depend on having even stronger and more widely accepted standards. The
Grid environment represents the future promise of even more open standards
and protocols that allow users to access the extensive data and powerful
processing, and combinations of computing resources worldwide that cut
across administrative and political boundaries.
REFERENCES
ArcNews Online. 2003. “GIS Standards and Interoperability.” ArcNews Online,

Spring. Redlands, CA: ESRI Inc. Available at www.esri.com.
ArcNews. 2005. “DS Waters Increases Efficiency and Profits with GIS and LBS.”
ArcNews, Summer. Available at www.esri.com.
Beitz, David Z. 2001. “GIS Assists Site Location in the E-Commerce Age.” Proceed-
ings of the 2001 ESRI User Conference. Redlands, CA: ESRI Inc. Available at
www.esri.com.
Buyya, Rajkumar, and Srikumar Venugopal. 2005. “A Gentle Introduction to Grid
Computing and Technologies.” CS Communications, 29(1): 9–19.
Chen, Anne. 2005. “MasterCard Cashes in on Web Services.” eWeek, February 7.
Available at www.eWeek.com.
Davis, Fred D. (1989). Perceived usefulness, perceived ease of use, and user accep-
tance of information technology. MIS Quarterly 13(3), 319–340.
Davis, Fred D. (1993). User acceptance of information technology: System char-
acteristics, user perceptions, and behavioral impacts. International Journal of
Man-Machine Studies, 38, 475–487.
Dhillon, Gurpreet. 2007. Principles of Information Systems Security. New York: John
Wiley and Sons.
Dodd, A.Z. 2005. The Essential Guide to Telecommunications, 4th Edition. Upper
Saddle River, NJ: Prentice Hall.
ESRI 2005. “GIS Standards and Interoperability.” ArcNews, Spring. Available at
www.esri.com.
FGCD. 2006. Federal Geographic Data Committee Website. Reston, Virginia: FGCD.
Available at www.fgcd.gov.
Foster, 2001. “The Anatomy of the Grid: Enabling Scalable Virtual Organizations.”
The International Journal of High Performance Computing Applications, 15(3):
200–222.
Hagel, John. 2002. “Edging into Web Services.” McKinsey Quarterly, November,
5–13.
Ismail, Ayman, Samir Patil, and Suneel Saigal. “When Computers Learn to Talk: A
Web Services Primer.” McKinsey Quarterly, November, 15–21.
Lopez, Xavier R. 2004. “Location-Based Services.” Chapter 6 in Karimi, Hassan,
and Amin Hammad (eds.), Telegeoinformatics: Location-Based Computing and
Services, Boca Raton, FL: CRC Press.
Maguire, David J. 2005. “Enterprise Geographic Information Servers: A New In-
formation System Architecture.” In Proceedings of the Annual Conference of the
Geospatial Information and Technology Association (GITA), GITA: Aurora, CO.
Markoff, John, and Martin Fackler. 2006. “With a Cellphone as My Guide: Digital
Search Meets the Real World in the Streets of Japan.” New York Times, June 28,
p. C1.
Miller, Alex. 2005. “Business Panel Discussion. Seaspan Case Study.” Presentation
at ESRI User Conference, San Diego. Redlands, CA: ESRI Inc.
Open Geospatial Consortium. 2005. Open GIS Specifications. Wayland, MA: Open
Geospatial Consortium. Available at www.opengis.org.
Puckorius, Timothy J. 2005. “Will Virtual Earth Scale Up the Spatial Market?”
Geospatial Solutions, 15(9): 30–33.
Ray, Julian J. 2005. “A Web-based Spatial Decision Support System Optimizes
Routes for Oversize/overweight Vehicles in Delaware.” Decision Support Systems,
in press.
Reed, Carl. 2004. “The Spatial Web.” White Paper. Wayland, MA: Open Geospatial
Consortium. Available at www.opengis.org.
Reed, Carl. 2004. “Integrating Geospatial Standards and Standards Strategies into
Business Processes.” White Paper. Wayland, MA: Open Geospatial Consortium.
Available at www.opengis.org.
Seaspan. 2007. Seaspan Ship Management website. Available at www.
seaspanmanagement.com.
Senne, Steven. 2005. “New Computers Make Grocery Carts Smarter.” USA Today,
May 3.
Stankiewicz, Amy. 2005. “A View from Above.” Geospatial Solutions, 15(7): 30–35
Zhao, Peisheng, Genong Yu, and Liping Di. 2007. “Geospatial Web Services.” In
Hilton, Brian (Ed.), Emerging Spatial Information Systems and Applications,
CHAPTER 6
SPATIAL SYSTEMS DEVELOPMENT
INTRODUCTION
The book has considered multiple ways that GIS and spatial applications
are benefiting businesses and organizations. Cases and other examples have
demonstrated these advantages, such as Sperry Van Ness in commercial real
estate and Chico’s in customer relationship management. The focus has been
on the features of contemporary spatial systems and how their applications
are helping businesses and their customers. However, another side of the
entire GeoBusiness field concerns how those applications are designed and
developed. This chapter’s central theme is designing and developing spatial
systems, rather than operating them. Some of the content of this chapter draws
from the fields of information systems and project management, which go
back further than GIS. The chapter builds the central theme by first examin-
ing the frameworks and methods for planning and designing spatial systems.
It discusses the well-known system development life cycle (SDLC) applied
to GIS and spatial techniques. The small but growing knowledge of spatial
design is included. The implementation phase consists of building, testing,
and activating the GIS application. Frequently development teams will de-
cide to outsource implementation to external parties. The alternative type
of end user development of professional and scientific spatial applications
is discussed. The outcomes may be decision support systems, software for
R&D, or professionally-oriented models. An example is a epidemiologist in a
pharmaceutical research center who produces spatial models and mapping of
disease diffusion. End users are highly skilled and can create spatial models
169
170 SPATIAL SYSTEMS DEVELOPMENT
that significantly help in decision support or discoveries. However, most of

the chapter focuses on formal development of business spatial applications by
the GIS and IT groups in organizations that use formal development methods
and project procedures.
The case study of a small GIS consulting firm, Engineering Systems in
Los Angeles, focuses on how the firm follows these system development
steps as an outsourcer in working with client users who are mostly from local
government.
SYSTEMS DEVELOPMENT FOR GIS
A GIS system must be conceived, planned, and taken through a series of

analysis, design, testing, and implementation steps to actually reach the point
that it is “turned on” and put into use. Many problems and mistakes can occur
along the way that might lead to a faulty system, the wrong system, or a
system that works but is too expensive. A major tenet coming from both the
IS and business-planning disciplines is that of systems development phases.
It is commonly accepted that a GIS can be better planned if the process is
divided into steps. A standard set of systems development phases applied to
GIS is presented in Figure 6.1. A brief discussion is given now of the phases,
and a more detailed discussion follows.
Figure 6.1
SYSTEMS DEVELOPMENT FOR GIS 171
The Planning phase involves conceptualizing the system, doing broad

planning, and justifying the need for a GIS. It ends by the decision of top
management on whether or not to commit to the project. The Analysis phase
emphasizes discovery and information gathering, leading to identifying a set
of alternatives. Feasibility and cost-benefit are weighed before prioritizing
a list of alternatives that is presented to top management. For instance, a
large insurance firm is analyzing what GIS would be best to implement for
underwriting and environmental modeling applications. It needs to gather in-
formation on the sources of data, types of architectures (client-server, web),
and types of analytical tools to solve the problems. In the process, it discovers
a third GIS use for the firm—marketing. It determines four alternatives to
be feasible and beneficial, and presents them in priority order to top man-
agement for an executive decision on whether to go forward with one of
them.
In Design, the system and data specifications are worked out in much more
detail. It involves diagramming, researching, and specifying of the fine points
of the prospective system. Users are extensively consulted on details of what
is being designed. Implementation is the phase of actually building the spatial
application, based on the detailed design. The design group has a key decision
to make here of whether to build it in-house, purchase it commercially, or
outsource it. In this phase, the in-house and customer users are trained in how
to use the system. The actual cutover to the new spatial system takes place
and the system is turned on.
In the Maintenance phase, the system is up and running, but users are
continuing to have issues and problems including training, correcting bugs,
and requesting features be added that were left off the original system. These
problems are addressed, often over many years. The final disposition of
the system involves its abandonment or replacement. In the latter case, the
replacement is the result of a new systems development process.
The systems development phases are iterative. In other words, at any time
during development, it may be necessary to return to an earlier phase or
phases and go through the phase(s) again. For instance, as seen on the right
of Figure 6.1, a new RFID system for inventory may have had a fine design
and the system was built well, but the implementation was botched because
the users were mostly not trained at all, and resisted the new system as
being too difficult. Using the iterative method, the SDLC can return to the
implementation phase, focusing this time on success in the sub-phase of
training, and can complete this phase the second time with turning on the
revised system for now well-trained users.
An alternative set of phases for GIS system planning by Tomlinson (2003)
comprised ten stages (see Table 6.1). The book’s systems development phases
are compared on the right side. The work by Tomlinson focuses mostly on
planning, analysis, and especially the design phase. It is informed by practical
experience in implementing GIS systems and is referred to in discussing those
phases.
TABLE 6.1 Tomlinson’s Ten-Stage GIS Planning Process, Compared to Systems

Development Phases
Tomlinson Planning Stages Systems Development Phases
1. Consider the strategic purpose 1. Planning

2. Plan for the planning
3. Conduct a technology seminar 2. Analysis
4. Describe the information products
5. Define the system scope
6. Create a data design 3. Design
7. Choose a logical data model
8. Determine system requirements
9. Consider benefit-cost, migration, and risk analysis
10. Make an implementation plan 4. Implementation
5. Maintenance
Source: Tomlinson, 2003.
Methodologies, models, and tools are available from the information sys-
tems, economics, and software engineering professions and disciplines that
can be useful for GIS development (Satzinger et al., 2007). Structured design
is a well developed system and software design method that advocates break-
ing the system into modular parts. Sometimes the modules are arranged in
a hierarchy known as top-down design. It has the advantage that the whole
design and high-level steps can be better understood. Associated with struc-
tured design is structured programming, which is an approach to computer
programming of breaking a program into modular sub-programs. The sub-
programs can be better focused and also can be re-used, to save time.
A high-powered, software-driven design method known as Computer-
Assisted Software Engineering (CASE) is implemented by using software
tools known as CASE tools. CASE tools support storing common metadata in
a data dictionary, a repository for the entire system’s development team. The
CASE tools also have features for diagramming the processes, data flows, and
data modeling of a system. Among the well-known diagrams are data-flow
diagrams (DFD) that show processes and data flows, entity relationship di-
agrams (ERD) that portray the relationships between data entities, structure
charts that give the big picture of object-oriented design, and class diagrams,
that show the hierarchical relationships of object types.
Another category of methods includes rapid application development
(RAD) and rapid prototyping. In RAD, the system development phases are
performed more rapidly. They can be performed simultaneously and/or it-
eratively, rather than being done sequentially. RAD favors CASE tools and
prototyping to speed up the steps. Prototyping refers to moving very quickly
to produce a limited final product for the user to test and critique. It has the
advantage that mistakes and errors in design can be picked up by the user,
SYSTEMS DEVELOPMENT FOR GIS 173
fed back, and improved quickly. The final prototype, once approved, needs to
go through some or all of the more formal systems development phases for
thorough testing and finish prior to implementation.
The case study firm Motion-Based Technologies used a combination of
RAD and prototyping to build its web-based spatial applications for high-
performance athletes. As a start-up, the firm’s experienced web programmers
wanted to move quickly. They could do so since the small firm had an informal,
flexible atmosphere without systems project procedures and controls. The
feedback was limited since the firm was small, so internal improvements
have come mostly from live users.
For the project management dimension of systems design, project manage-
ment software is utilized that can track the tasks, workers, costs, and critical
pathways for a project. Cost-benefit methods, explained in Chapter 7, such
as ROI and net present value, are utilized in the planning phase as well as to
measure actual net benefits in the maintenance phase.
The actual use of the systems development methods and tools for GIS
and spatial technologies is limited in industry, as will be shown later by
reviewing the twenty research cases. This may be due to: (1) the history
of GIS development occurred mostly in the public sector which is not as
well versed in the techniques; (2) in a company, the GIS design team is often
located in a small independent group with members drawn from non-business
backgrounds; and (3) fewer big-corporation design methods and techniques
are imposed for GIS, since GIS is rarely a visible profit center for larger
companies. The positive aspect is that nearly all of these development tools
apply to GIS and are widely available and documented extensively (Satzinger
et al., 2007).
There are two major approaches to systems development—the tradi-
tional approach and the object-oriented approach (Satzinger, 2007). Both
approaches go through the five phases that have been outlined. The tradi-
tional approach is based on structured design and structured programming.
Two of the major diagrams utilized are the data flow diagram and the entity
relationship diagram. The problems with the traditional approach often con-
cern lack of consistency in explanations of it. Also, it is difficult to use for
modeling highly object-oriented solutions.
Object-oriented (OO) system development takes a very different view. The
key item is the Object, which is defined as an item in the computer system
that has attributes and behavior and is capable of sending and responding to
messages. In object-oriented spatial systems development, a set of interacting
objects is designed to work together to accomplish goals. The object replaces
the processes, programs, and data entities from the traditional approach at the
application level, although individual objects have more focused program-
matic elements and use of data. The OO approach has the advantage that
the object is easier to think of as relating to the real world. Also, objects
can be re-used in different programs more easily than for traditional pro-
cessing procedures. For instance, a firm has the spatial objects for customer,
product, and store that are reusable, naturally meaningful, and can include
location.
Either the traditional or the object-oriented approach can be suitable for
designing spatial systems. The traditional is more appropriate for systems that
are slowly changing and have complex geographical layering, while object-
oriented is better for more rapidly changing, interactive systems with simpler
layering. The phases of systems development apply for both approaches, and
the discussion below does not distinguish between the two.
PHASES IN SYSTEMS DEVELOPMENT FOR GIS
This section goes into more detail on each phase.
The Planning Phase

In the project planning phase, the spatial problem or opportunity that needs
a solution is identified. The phase then defines it further and looks at its
feasibility, budget, staffing, scheduling.
Recognition of the problem may derive from management, professional
GIS or IS staff, users, or customers. For instance, a retail executive in a mid-
sized firm is concerned that his stores are not located in the most rapidly
growing market areas. At a meeting, his IS manager mentions that GIS is
used by some competitors to identify growing markets by small-area. Another
illustration is drivers in a small trucking firm, who ask their management for
a sophisticated GPS-based truck navigation system.
Once identified, the problem must be checked for feasibility and budget.
Can it be accomplished technically within the capabilities of the organization
and/or providers? Is it affordable? An example of these issues is seen with
ABC Blinds Inc., the fictional name for a $2 million–revenue small business
in California. The firm has the idea to implement GIS for marketing using a
commercial GIS package, but key questions remain: Can its single outside
technical-support person understand and apply GIS for the marketing problem
at hand? Can ABC afford mainstream commercial GIS software or services?
The next substeps are to determine staffing and scheduling of the project.
Scheduling needs to take into account project difficulty, sub-tasks, availability
of personnel, and the critical steps necessary for project completion. Staffing is
a challenge for most organizations anticipating GIS projects. This challenge
is partly the sectoral one of “knowledge gap,” i.e. a situation in the labor
market where a technology advances faster than the capability of businesses
to find staffing for it (Tomlinson, 2003). This has applied generally to the
GIS field over the past ten years (Marble, 2006). One response would be to
allocate budget and time for training of internal staff. At the societal level,
the knowledge gap points to a need for universities to build up more GIS
programs across a variety of disciplines (Marble, 2006).
PHASES IN SYSTEMS DEVELOPMENT FOR GIS 175
GIS Systems Development by Professionals and Scientists
For the GIS field, the staff who design, build, and implement spatial systems come
from a variety of backgrounds that relate to the history of GIS. From its early days,
GIS has been interdisciplinary, drawing in people from geography, environmen-
tal sciences, architecture, landscape architecture, computer science, engineering,
public administration and government. Some of these users are engaged in sci-
entific and professional spatial work, such as in corporate R&D centers, univer-
sities, and government research organizations. They contrast with personnel for
the growing GIS markets in the private sector, who come from such fields as busi-
ness, management, marketing, information systems, and geographic information
sciences.
Systems development for the scientific users falls into the category of “end user
development.” Such users still accomplish the stages of systems development and
use some of its methods such as prototyping and software testing, but they proceed
as individuals or small groups through less structured and often iterative steps,
and lack the formal controls of a professional development team. For example,
a faculty ecologist at a university has developed a system to produce thousands
of maps of ecological habitats worldwide. The mapping system is controlled by
object-oriented programs that technical staff in her lab developed in consultation
with the university’s scientific programming unit. The ecologist is expert in spatial
modeling of habitats and draws on lab assistants who are have become trained
and experienced in GIS software and programming. The users for the software are
eight scientists and graduate students, who produce scientific reports and papers.
This example illustrates an important type of systems development, but one
that tends less to follow the formal phases and controls mostly examined in the
chapter. This professional or scientific GIS user often has substantial knowledge
of GIS software, models, data, and tools. Their systems tend to be at the level of
DSS. Today, there continues a significant community of such professionals, who
contribute important spatial discoveries and insights. However, in the corporate
world today, GIS staff tend more to be business, IS, and applied GIS specialists,
who develop transaction-processing, MIS, and enterprise systems that serve larger
groups of users under tighter performance requirements. It is this latter group that
utilize the formal development approaches and methods discussed in most of this
chapter.
The training challenge for a business applies across all levels of GIS and
spatial workers and users, starting with the GIS manager and including staff,
users, and customers (Tomlinson, 2003). Since the technology is moving
fast, even an experienced GIS manager must plan for his/her own training.
Training deficits can undermine the system later in development. “One of
the primary causes of under use of GIS is lack of staff training” (Tomlinson,
2003). At the worst, not sustaining skilled GIS staff or outsourced capabilities
can eventually cause the company to regard the GIS as a failure and abandon
it. Training also is important for end user developers, but is more specialized.
The planning phase culminates in a proposal to top management to go

forward with a proposed project. If approved, the team commences work on
the project.
The Analysis Phase

Analysis involves information gathering, discovery, visiting and hearing the
users’ needs, developing the spatial system requirements, prioritizing the
requirements, and assembling a proposal for full design and implementation.
A systems development team is assembled for this project, to take it from
analysis all the way to an implemented solution. The project team is usually
composed of a team leader, GIS and IS analysts, and a technical person who
will play a greater role later on. The team leader might be the company’s GIS
manager, but often is appointed by that manager. Analysis is varied as seen
by the multiple approaches used by the case companies.
One example is Baystate Health, where the staff of four GIS professionals,
headed by the GIS technical manager, constitutes the spatial development
group. The professionalism of a GIS staff is rare for hospitals. This technical
group also has as its director a surgeon and chair of Baystate’s Surgery
department, who provides organizational leadership and lends subject-matter
expertise to some of the GIS applications. The staff uses both formal and
informal systems development approaches. The formal approach is helped by
one staff member who worked for ten years at a leading GIS software vendor
firm, which required formal protocols.
In Analysis, information is gathered in different ways depending on the
project type and its needs. Frequently, if there is an existing system, it is
analyzed and evaluated. It is run, tested, and its pluses and minuses evaluated.
This allows weaknesses to be addressed in the requirements for a new system.
A variety of stakeholders are surveyed, questioned, and consulted on their
concerns and advice, including top executives, middle managers especially in
user departments, users inside the company, and external customer users. This
step is well known to be perhaps the most crucial (Laudon and Laudon, 2006;
Satzinger et al., 2007), since the present users and customers will be among
the most important eventual judges of the implemented spatial product.
Information is gathered from other diverse sources, including industry and
vendor literature, consultants, other companies that are not directly compet-
ing, and websites that provide spatial services. The composite of this informa-
tion and knowledge is not just stored, but is analyzed, synthesized, and prior-
itized. For example, a middle-sized private home security firm intends to up-
grade the firm’s GIS and GPS navigation systems for its fleet of agent vehicles.
It has interviewed its agents for their critiques of the existing system, received
an expensive consulting survey on the pluses and minuses of commercial sys-
tems, and received detailed feedback on the candidate system from a user
firm in another country that is not competing. The project team goes through
this mass of information, and connects together themes and conclusions,
discarding some information as unreliable. It distills the mass down to eigh-

teen prioritized key points.
It is useful in analysis, if time and resources allow, to build a small proto-
type of the most essential part of the new system. This prototype will show
dramatically where there are weaknesses in the project and its requirements.
Better yet is to involve users in reviewing a prototype.
URS Corporation, a large integrated, engineering design services company
with a global reach, serves as an outsourcer to develop its clients’ government
and environmental GIS systems. URS GIS specialists develop specifications
and design, following by a prototype. To the user, the prototype resembles an
actual system minus error-trapping and finished data-base structure and func-
tions well to serve workflow. However, it has “dummy” data. The client tests
the system and gives more requirements, a process that may go through sev-
eral iterations. There may be creep in the project scope. Only after agreement
is reached with the client on the prototype is the finished system built.
A variation of early prototyping for GIS (Tomlinson, 2003) is termed a
“Technology Seminar” (see box.).
The Technology Seminar for GIS Planning
Tomlinson (2003) proposes to hold a “Technology Seminar” very early in planning

a GIS project. This concept is not too different from Joint Application Development
(JAD), which is a group-meeting technique often used in systems development for
middle- and large-sized organizations (Laudon and Laudon, 2007; Satzinger et al.,
2007). A Technology Seminar is called by the GIS manager early in a project with
the goals to provide some training, to increase understanding of GIS and spatial
applications, and to have the group develop recommendations for requirements of
a GIS product or products. The size of the Technology Seminar group ranges from
a dozen for small organizations to thirty or more for large ones (Tomlinson, 2003).
The group consists of key GIS-user department heads, some key staff users, the
CIO, some middle managers, and a few executives.
The agenda for a Technology Seminar is the following (Tomlinson, 2003):
r GIS definition
r GIS terminology
r Functions of a GIS. This might include demonstrating examples of spatial
hardware and software for exposure and training, not giving bias.
r The planning process: steps and responsibilities
r Initial identification of first requirements for the new or updated GIS system
The key question for such a Technology Seminar is what information is needed
for you from the spatial system? (Tomlinson, 2003). The information may not be
a map, but could be lists distilling the outcomes of spatial analysis. When the
group coalesces on a project deliverable, it can be identified and recorded along

with its requirements and the name of the principal person who proposed it. The
session format precludes a person’s manager from overruling or influencing the
proposals from his/her staff. After the session, the project deliverables are checked
to make sure they are in synch with the company’s strategic plan and then they are
prioritized. Another key question afterwards is, which user(s) or user department(s)
will benefit? It is recommended that a Technology Seminar be held at the very
beginning of analysis, and experience has shown it to be beneficial in discovering
what the user really needs (Tomlinson, 2003).
Another key part of analysis is to recommend how the GIS application

will be built and by whom. The alternatives are (a) build it in-house with staff
from the GIS and/or IS departments, (b) build it through end-user development
by high-level professionals or scientists located in user departments of the
organization, (c) purchase the software commercially, and (d) outsource part
or all of the construction to an outside provider.
Building in-house (a) requires that a team of technically skilled program-
mers, software engineers, or GIS specialists be available or be hired. In end
user development (b), a skilled expert develops and builds a specialized ap-
plication. This might involve spatial modeling, statistics, scientific program-
ming, or decision support. End user development often involves less formal
development steps, high analytical skills, and products that have a profes-
sional or scientific set of users. Purchase of the GIS software (c) can lead to
quick implementation, but careful study is required to ensure that there is a
good fit with user needs. The outsourcing alternative (d) reduces the technical
burden, but often involves higher costs for software, hardware, training, and
maintenance. Sometimes, this choice applies more broadly by not only having
the outsourcer do construction but also get involved with analysis and design.
Outsourcing has become more common for spatial projects, but has pluses
and minuses that are covered more in Chapter 10. Analysis ends by the project
team presenting its top recommended alternative, along with justification and
high-level requirements to top management for its review and approval.
Design
In this phase, the complete design and full set of requirements are developed
and created. It includes designing the architecture of the application, physical
hardware including networking, design of the data model, design of the spatial
application, identifying the best sources of digital data, checking on digital
standards, planning data conversions, detailing the tabular, graphical, and
spatial outputs, and outlining the spatial cartography. These tasks combine
GIS, IS, and business knowledge and skills, often blending them together.
Organizationally, it involves communication between staff in the three areas,
Figure 6.2 Data Flow Diagram for Marketing Map.
and better yet, staff from all three areas working together on the design team.
For a small firm, or in an outsourcing situation, having every area represented
on the team may not be feasible.
The architecture design for the spatial application often is expressed by
diagrams, charts, and tables. For the traditional approach to systems devel-
opment, well-known design diagrams are the data flow diagram (DFD) and
entity-relationship diagram (ERD). The data flow diagram (see Figure 6.2)
shows the components that are sending and receiving input and output (square
symbols on the figure), the processing steps (squarish symbols with the curved
corners) and data stores (C-shaped symbols), and the data flows (lines with
arrows).
The process represented in the figure is the building of a three-layer mar-
keting map to be output to the customer. The map is being built by XYZ
Marketing Consultants. There are four paths vertically on the diagram, the
left two producing two of the layers, and the right two paths produce the third
Figure 6.3 Entity-Relationship Diagram Example.
layer. For instance, the left column gathers the customer address list, stores it,
digitizes it to ZIP codes, and produces the map layer showing customer point
locations. At the end the finalized three-layer marketing map is produced for
the customer.
The DFD can represent the spatial and non-spatial processing and storage
steps. For instance, one data store holds the “customer address list,” which
is non-spatial, while another one stores the “ZIP Code digital boundaries,”
i.e., the spatial data. Although not shown here, the DFD is also capable of
portraying the processing and storage of combined attribute and spatial data,
such as happens with the object data model (see Chapter 8).
The DFD method has the advantage that any processing symbol, for ex-
ample, “Digitize Customer Addresses by ZIP Codes,” can be “exploded” into
a whole DFD diagram with its own more detailed process, data flow, and
storage symbols, which is located one level below. If all the process symbols
at this level were exploded, there would be nine more detailed diagrams at
the level underneath that shown in Figure 6.2. The process of developing a
multi-level DFD is available to readers (Valacich et al., 2004; Satzinger et al.,
2007).
An ERD shows the relationship between data entities, such as “Customer,”
“Store,” and “Food-Item” (see Figure 6.3). Each entity has attributes. In
some cases entities can be spatial, for example the attributes of point location
(Latitude, Longitude or other X, Y coordinates), line coordinates, polygon
coordinates, and other spatial units, etc. The relationship between entities A
and B may indicate for each A there is one B (1 to 1); for each A there are
many Bs (1 to n); there are many As for each B (m to 1); or there are many As
for many Bs (m to n). These relationships are known as cardinalities. They
can be helpful in design of spatial systems.
In the figure, the entity “Store” is spatially referenced to a point by Latitude
and Longitude. It also indicates its relationship to the spatial category of ZIP
Code. The other entities, “Customer” and “Food-Item,” are not spatially
referenced. The cardinalities are indicated by the labels on the connecting
lines. Customer and Store are M:1, while Store and Food-Item are 1:N.
DFDs and ERDs constitute robust diagramming techniques that can be
applied broadly to design spatial systems. They are also available as features
in CASE tools. The multi-layer aspect of DFDs, and the connections between
DFDs and ERDs can be automatically linked in a CASE tool. In the alternative
Object-Oriented Systems Development Approach, there are other types of
diagrams not covered here (Satzinger et al., 2007) that may be more or
less useful than the traditional approach for modeling spatial development,
depending on the type of problem.
The physical design of the spatial system is another major part of this
phase. It includes the following components:
r Computing hardware and performance

r Network design and performance
◦ Internet and web design and protocols
r Spatial hardware
◦ Digitizers, high-resolution printers, plotters
◦ GPS, RFID, and sensor hardware
◦ LIDAR hardware and other airborne hardware
r Mobile devices
◦ Cell phones
◦ PDAs
◦ Tablet and pocket pcs
r Spatial software
◦ General purpose commercial software
◦ Database software
◦ Software for spatial hardware and mobile devices
◦ Custom application software
r Capacity planning for future expansion of hardware base
r Backup hardware
r Plan to interface hardware with the corporate network including enter-
prise systems
The physical design needs to support the logical design. It should be

carefully analyzed to confirm that all of its parts are compatible with, and
connect well with each other and with the organization’s existing physical
infrastructure. An example of the challenge is a small delivery company that
acquired GPS units for its fleet of vehicles, only to find out too late that it was
difficult to import the GPS coordinate files collected in the field back into its
commercial GIS software. It had to hire a consultant to write software that
accomplished this conversion. The firm’s logical design was fine, but at first
it did not work with the physical devices.
Design for the acquisition of spatial data is essential. Digital sources of
spatial data are quite varied and include government, commercial vendors,
and in-house. Government data are often free or nominal in cost for the U.S.
but can be quite expensive in European and other countries. Commercial data
vary in price and quality including the extent of metadata and documentation
available. Government and commercial data sources are described and often
can be tested at spatial data portals, such as the U.S. Census Bureau (www.
census.gov), ESRI’s Geography Network (www.geographynetwork.com), or

the U.S. government’s Geospatial One-Stop (www.geodata.gov). In-house
data are often proprietary spatial data that have been digitized, collected by
GPS, or collected and scanned from airborne systems.
For example, Kaiser Permanente, the California-based healthcare firm, ex-
amined in Chapter 10, applies GIS for travel-time analysis, site selection,
regulatory reports, a library of reference maps for members posted on in-
tranet, spatial web services for members, support for an ambulance service
in the Bay Area, epidemiology research studies, sales and marketing, and
finance/accounting. For these applications, government, commercial, and in-
house data serve as input to the GIS. The finance/accounting data and epi-
demiology data are proprietary and in-house. The travel-time analysis is based
on street-map data from commercial vendor TeleAtlas and the PEMS data
on freeway performance measurement from a state government-university
consortium (PEMS, 2007). The data for map services for members come
from a commercial service. The data for the intranet library of reference
maps are mostly in-house and proprietary such as member locations, but
utilize some government data. Kaiser’s GIS is not put in one central repos-
itory but resides in many locations keyed to specific projects throughout
the firm.
For any spatial data, its characteristics are critical to examine. As seen in
Table 6.2, the most important factors are scale, resolution, map projection,
and error tolerance (Tomlinson, 2003). Map scale is critical. First of all, the
TABLE 6.2 Criteria to Evaluate Spatial Data in System Design

r Scale.
The map scales need to be not too large, which makes data storage unacceptably large, and
not too small which lowers resolution and weakens the business purpose. Scale needs to be
compared so map layers have the same scale and remain accurate.
r Resolution.
Resolution is closely tied to scale. It shows the smallest map features that can be identified
for a certain scale. Thus it needs to be fine enough to satisfy the business problem solving
with the given map.
r Map Projection.
The map projection of each map needs to be known and accurate. There needs to be a check
done that the conversions can easily be done into a common projections for all the layers on
a map.
r Error Tolerance.
The error tolerance is an estimate of the maximum error that can be tolerated for particular
business problem. Once the tolerance is determined, the project team needs to estimate the
cost to achieve that tolerance. There are methods available to analyze the major error types
and how they relate to the problem at hand, in order to estimate cost.
Source: Tomlinson, 2003.

extent that map scale is limited in accuracy is based on the scale and accuracy
of original map source. The current map cannot have larger scale or greater
accuracy. If scale for a certain application needs to be small to conform to
other layers in a map, a larger scale map can be stored for separate display
(Tomlinson, 2003). The map resolution needs to be considered relative to the
business problem. For instance, a thematic map portraying levels of market
penetration in eight national sales regions can get by with low resolution,
whereas a map that shows where utility gas lines cross under residential
properties must have high resolution.
Design of data standards and data conversion depends on the sources of
data and what is needed internally by the planned system. There are over thirty
GIS data interchange formats (Tomlinson, 2003). Among the most important
are .DLG (digital line graph), TIGER (from U.S. Census Bureau), and .DXF
(the interchange format between CAD and GIS systems). The project team
needs to consider if the in-house GIS software will accurately and efficiently
convert data to the correct internal format from data being imported by one
of these standards. The challenge of data standards and conversions extends
to associated spatial technologies, such as RFID, sensors, and data being
transferred over the internet and web. The book covered the major spatial
web standards in Chapter 5, but it’s beyond the scope to cover the multiple
standards and conversions. The main point is that the spatial project team an-
alyzes and resolves standards, compatibilities, and data conversions as part of
design.
Design for controls and security has the goal to assure the system has in-
tegrity and safety to an appropriate and affordable level. GIS security, covered
in Chapter 11, involves strengthening the technical security, as well as the for-
mal and informal controls. The techniques of doing this range from passwords
and firewalls to hiring practices and provisions for security education and
training.
Prototyping can be applied to very quickly design, build, and test parts of
the system or a whole system. Design team member and users can evaluate
system functionality and performance. Some tools are available for spatial
prototyping. For instance, ModelBuilder by ESRI (ESRI, 2000) has tools to
allow rapidly building a model by adding graphical symbols to a flow diagram
(see Figure 6.4), and testing them with real data (see Figure 6.4). It can be
utilized during design to test the feasibility of a business spatial model.
Cartographic design focuses on cartography, the art of giving appealing
and readable appearance to maps which was mentioned in Chapter 2. GIS
software packages have built-in cartographic frameworks, but they are not
high-level cartography keyed to the particular GIS problem at hand. Carto-
graphic design can be improved by having a design-team or consulting car-
tographer add requirements for cartography. Some software packages allow
the cartography to be replicated for an entire project. For example, ESRI’s
geodatabase feature allows a cartographic format to be entered which will
Figure 6.4 Copyright © 2007 ESRI. All rights reserved. Used by permission.
enforce consistency across all maps in the project, even new ones. In short,
the cartography can be specified as part of the design.
The Implementation Phase

In Implementation, the system is built, tested, and put into use. The spa-
tial system is constructed, based on all aspects of detailed design. This can
be done in-house or outsourced. Spatial and attribute data are acquired and
loaded into databases, data warehouses, or directly into tables in the GIS
software. For this substep, some data conversions may be needed. Hard-
ware and networking installations or upgrades take place and the entire
physical system is connected together. Other spatial technology compo-
nents including GPS, RFID, sensor, mobile devices are acquired, installed,
and tested.
URS Corporation, discussed earlier, gets first an OK on a prototype and then
builds the finished system. For implementation, the team runs and extensively
reviews and critiques the finished system on its server. Next, the tested system
is put on the client’s server host in Beta version for the client to test and request
CASE STUDY FINDINGS ON SYSTEMS DEVELOPMENT 185
any final changes. Only after this is the system implemented for the client’s
users.
The internal users and customers need to be trained. It can be done by web-
based training, in-house or external training workshops. It is a critical factor
for success, since users’ responses ultimately determine system success. As
mentioned, users may be unfamiliar with GIS and spatial concepts, so they
need extra time and attention (Tomlinson, 2003; Marble, 2006).
The new system is “turned on.” For replacement of an existing system, a
process of cutover from the old to the new system takes place. The old system
sometimes runs concurrently, which gives users reassurance that they can fall
back on the old system if necessary.
The Maintenance Phase

The final phase supports the smooth running of the spatial system for years fol-
lowing the implementation. It includes the following: (1) maintaining smooth
system functioning, (2) adding enhancements, (3) debugging errors discov-
ered after implementation, and (4) providing continuing training and support-
ing the users (Valacich et al., 2004; Satzinger et al., 2007).
For users, this phase opens up the benefits and productivity gains of the
new application. On the other hand for developers, maintenance may be dull,
punctuated by sometimes ugly system repairs and debugging. It has been
known for decades that job motivation of maintenance staff is among the
lowest of systems professionals (Couger and Zawacki, 1980). The manager
of GIS needs to foresee this problem, and attempt to offset it with incentives for
maintenance staff, regular rotations in and out of maintenance, or enhancing
its work.
The Maintenance phase ends in one of several ways. The GIS or spatial
system can be replaced through implementation of a new system stemming
from the development process. Referring to Figure 6.1 on the right, there
is a loop back to start a new development cycle beginning in the Planning
phase. It goes through all the phases leading to the new system. As seen in the
diagram, the looping back may only go back several phases, in which case the
new system is referred to as an upgrade. Another ending to Maintenance is
through abandonment of the system. It might occur due to a shift in company
strategy away from spatial technologies, or as an offshoot to corporate merger
or restructuring.
CASE STUDY FINDINGS ON SYSTEMS DEVELOPMENT
This section next turns to the book’s research case studies to assess what
systems development approaches are actually used in the real world by the
sample of twenty companies. The sample of 20 provides findings on the

following:
1. Was systems development done in-house or mostly outsourced?

2. If in-house, were formal phases taken, or were informal steps followed?
3. What was the composition of the systems development project team?
4. Was the system development approach traditional or object-oriented?
5. Were there differences by the size of company in the approach taken?
Was systems development done in-house or mostly outsourced?

Regarding Question 1, only two firms of twenty had most of the development
done through outsourcing. The Large Credit Bank had very limited GIS
initiated. In the main credit bank, development was outsourced entirely to a
small firm and limited in scope to spatial analysis of the markets for branches
and mapping of the bank headquarters campus. The newly acquired banks
utilized some GIS already and sought more. It was as yet unclear if their further
spatial systems development would be outsourced. For LCB, outsourcing
appears to be the result of benign neglect of the spatial area, rather than
outsourcing for competitive benefits.
The second firm that mostly outsourced GIS development was Kaiser Per-
manente. Examples of major spatial systems development outsourced were
enhancements to the minimum-travel-time system, full design, and implanta-
tion of the service area mapping application. For the latter, outsourcing was
deliberately chosen by the GIS managers and supported by top management.
The justification was that the GIS department in-house consisted of only one
person, the manager who although talented as a designer and implementer
clearly was over-extended to conduct the bulk of development himself at the
scale of a middle-sized company. Instead the approach was for the GIS man-
ager to initiate development, get it on the right track through analysis, and
then turn it over to outside consultants to do the heavy design and program-
ming. Another route Kaiser could have gone would have been to build up a
skilled GIS staff reporting to the manager. However, the outsourcing approach
utilized has been generally successful, and underscores that outsourcing can
be worthwhile and beneficial, if planned carefully.
The other eighteen firms utilized primarily in-house development and im-
plementation. Although Sears contracted its ESRI GIS applications to outside
consultants, the bulk of all the spatial programming was done in-house by
the programming group in IT. Global Integrated Oil includes contractors on
some of its projects, but the bulk was done in-house by a combination of GIS,
IT, and business specialists. The Large Insurance Firm began its GIS project
design and programming with an outside contractor, was dissatisfied with the
results, and brought all GIS development work in-house. Two of the firms
in the sample, URS and Engineering Systems, primarily do in-house GIS
development acting as the outsourcer for other firms. Engineering Systems is

discussed later as the chapter case study.
If in-house, were formal phases taken, or were informal

steps followed?
For the eighteen firms doing in-house development, ten utilized formal sys-
tems development phases and methods. Their approaches can be examined in
Table 6.3. Some trends are evident: (1) The GIS team mostly leads in systems
development, sometimes partnered with the IS group. For Southern Com-
pany, IT leads, while for smaller firms of Sperry Van Ness and Engineering
Systems, GIS and IT are merged. This is not surprising given the specialized
knowledge for GIS. (2) The formal steps resemble the five phases discussed
in the chapter. (3) Three firms, Large Insurance Co., URS, and Sperry Van
Ness, include Prototyping as part of their formal procedures. It is not surpris-
ing since for different reasons, each of them had time-pressure to get things
done.
What was the composition of the systems development project team?

It’s less clear how systems development is done for the eight case companies
with an informal systems development process. For Prudential Preferred and
Lamar Advertising, there seem to be GIS staff or business staff who take a lead
in initiating ideas, some of which gather strength for development projects.
At Chico’s ideas come spontaneously from the small four-person GIS group.
The informal development has been successful for most of these firms using
it. It needs to be studied more to understand what makes it work well.
Was the systems development approach traditional

or object-oriented?
For firms using formal systems development approaches, they were all tra-
ditional. Motion-Based Technologies utilized informal steps for systems de-
velopment but for implementation centered on the Java language, making
use of its object-oriented features. One of the reasons for the dominance of
traditional is that systems development came mostly from the GIS group,
which largely lacked training or experience in OO development methods or
languages.
Were there differences by the size of company in the approach taken?

The formal approach is far less common for small companies (one-third),
compared to large and medium ones (60 percent). An explanation is that small
firms are at earlier maturity stages, are hard-pressed from day to day, and are
often short-handed, so formal procedures and controls are less prevalent.
TABLE 6.3 Case-Study Firms Using Formal Systems Development
Who does systems

Name Size development Steps taken (formal or informal)
Global L Groups within the company Development process depends on group

Integrated design, develop, deploy composition. There is a standard
Oil applications. internal project management process:
Combination of GIS, IT, (1) scope, (2) evaluate alternatives, (3)
and business employees build, (4) deploy, and (5) look back.
and contractors.
Large L Originally, GIS team Steps: (1) research with selected group of
Insurance outsourced design to underwriters reviewing good and bad
Co. external consulting firm. of original design. Over 1-year period,
Now it’s done in-house. GIS mangers and several others in GIS
group wrote it in faster format,
programming in C++. Used
prototyping. Implemented basic
functionality in one big application,
and later improved it over time.
Rand L GIS group leads if project Steps: (1) R&D, (2) Conceptual design,
McNally for internal use. If project (3) Requirements/prototyping, (4)
to be commercially Development, (5) Construction, (6)
packaged and sold, Rand Review, (7) Maintenance. Details vary
McNally’s programming project-to-project. IT group has more
group is the lead. rigorous requirements and tighter
quality control than GIS Group.
Sears L CAM application Development process involves a training
Roebuck development is done by program. Field users provide input in
the GIS Group, SST the requirements stage. Development
application by 24-person takes into account many details.
IT programming Prototyping and screen designs used.
management team, and
ESRI GIS applications
contracted to outside
consultants.
Southern L Usually developed by IT A “business driver” meets with a business
Company development unit in analyst and they together clarify the
Southern Company need. With more digging for details, a
Services. Sometimes business case is developed. Approval
developed in a subsidiary and funding OK are needed.
unit, but IT development Preference for commercial software
unit needs to “in the (“off-the-shelf”). If built in-house,
loop” systemwide. there is a highly disciplined approach.
Project manager who communicates
milestones.
TABLE 6.3 (Continued)
Who does systems

Name Size development Steps taken (formal or informal)
URS L The users are from client. URS GIS specialists work with client to
develop specifications. User needs to
know quickly how the application
works. A detailed design documents
gives design requirements early on
(“Quick and dirty straw man”).
Implementation. If hosted by URS
server, the prototype is run on the
server with client data. Beta version
includes “logins, error trapping, and
permissions for data access.” Everyone
on the project team reviews and
critiques it, and the Beta version is put
on the client’s server.
Arizona M Steps: (1) conceived based on user needs,
Republic (2) designed by DP staff in GIS group,
(3) developed by DP staff in GIS
group, (4) implemented. Prototyping
sometimes done.
Bay State M Staff of GIS professionals Staff uses both formal system design
Health are principal conceivers, methodology and ad hoc methodology.
designers, developers,
and implementers.
Sperry Van M Avoid the bureaucracy. Skilled prototyping approach. Build the
Ness Customize for a project first full-blown version in 3 months.
with particular players. Avoid 6–10 months of iterations.
Go to the customers of Steps: (1) determine what we want to
the applications. Ideas accomplish, (2) determine who uses it,
pour in from e-mail and (3) determine what you want to do with
marketing advisors will it, (4) go do it through prototyping, (5)
call in. roll it out, (6) adjust and fix.
Engineering S ES believers GIS is a For clients, needs and requirements
Systems significant component of analysis. Do a JAD session with senior
the IT system, and GIS person from each division. JAD team
and IT need to work meets monthly for 1–2 hours. Do
together. technical design based on business
needs, not just spatial data. How do we
help the sure meet business goals.
Develop the data model. Data
conversion. Detailed database design.
Implementation plan. Specifications
for each step. Budget specs.
Maintenance. Sometimes an Indian
outsourcer is used for data conversion.
In summary, the twenty research case studies help to understand the real-
world environment of systems development for GIS and spatial technologies.
Although it’s a limited sample, it suggests that most companies for GIS utilize
a traditional systems development approach, that large- and medium-sized
firms adopt formal phases for development, and that the GIS group plays a
major or the lead role in development, although the associated IT group tends
to conduct in-house programming and technical implementation.
CASE STUDY: ENGINEERING SYSTEMS
Engineering Systems (ES) is a small, Los Angeles–based GIS consulting firm

that employs fifteen people. Its primary customer base consists of local and
county government agencies in California and Virginia, with secondary em-
phasis on private companies. Examples of its clients include San Bernardino
County, City of Downey, and Metropolitan Water District in southern
California; and the Town of Vienna, City of Falls Church, and Fairfax County
in Virginia. ES provides GIS planning, development, and implementation
consulting services. The projects are usually multi-year, some as long as ten
years, with ES personnel often located at the client’s site. ES acts as a se-
lective outsourcer for GIS, and it works together with the staff of the client
organizations. For instance, for both Fairfax County and City, ES has worked
for several years to develop and monitor the quality of the GIS for Fairfax’s
IT services department, including mapping that shows the update and mainte-
nance of land parcel records (see Figure 6.5). Recently, ES has done a limited
amount of its work for high-tech spatial client companies, such as one that
helps manage supply chain logistics and AVL (automatic vehicle location)
for container boxes once they have arrived at seaports. Some of the work
requires highly accurate maps, for instance sub-meter accuracy for line maps
for utilities (gas, water, sewer and storm water).
ES has adopted a strategy to emphasize accuracy for all of its work. It
reasons that a new client that just wants to view low-resolution mapping of
tabular information may later require high accuracy. For instance, one very
large utility client initially did not care what side of the street utility poles
were located—it just wanted to see a schematic map. But eventually that
company realized that they required higher accuracy and the poles placed
within several feet of accuracy for their application to be functional. There
is, of course, a geometric relationship between accuracy and cost. Careful
design is required to analyze costs and benefits to optimize a system.
ES has successfully developed its own internal GIS system at its office in
downtown Los Angeles, but the bulk of its GIS development and the focus of
this case study is on the GIS development for its outside clients.
ES follows the standard steps of systems development, that resemble
the steps covered earlier. However, it always tries to start its development
CASE STUDY: ENGINEERING SYSTEMS 191
Figure 6.5 Engineering systems map of land panel update and maintenance for
Fairfax Country, Virginia.
engagement with a client with a Joint Application Design (JAD) session. JAD
is a systems development method that speeds up the gathering of the system
requirements. Instead of being spread out over months, JAD compresses this
gathering into several intensive meetings that include the client users, all the
key stakeholders, and the project team members. The participants are pulled
away from their normal daily activities, so they can focus on rapid sequences
of gathering facts, building system models, making decisions, checking on
policies, and verifying the results (Satzinger et al., 2007).
At the starting JAD session, ES invites a senior person from each division
or major department, who can make things happen during the session and
later on. Requirements gathering is done by drilling down deeper with each
division represented at the JADs. The JAD team meets every month if possible
to continue through subsequent design phases.
Design emphasizes determining the business needs and then performing
the technical spatial design based on those needs—not the other way around.
The data model is developed with an emphasis on the relational data model,
which helps to avoid data redundancies, a topic taken up more in Chapter 8.
A detailed implementation plan is developed along with detailed tasks and
budget specifications. Since most of the projects ES gets involved with are
large-scale, sometimes huge, the entire project may take up to ten years or
longer. Thus the project schedule is broken into five-year periods, with tactical
one-year objectives.
The largely government clients frequently have needed data conversions.
This is because data come in from a variety of external agencies and internal
departments. ES is today known for its expertise in data conversion.
However, ES has experienced its share of projects that failed. One large
utility client, one of ES’s first ones in the early 1990s, had invested $30 million
in a Geo-Info System (fictitious name) to try to integrate GIS information on
the utility physical network with its customer information system. Unfortu-
nately, the utility firm insisted on testing a prototype Geo-Info System on an
entire large county. Not enough focus was given to demonstrating payback
on the huge investment at the time. ES was called in at a late stage to try to
salvage a dying Geo-Info System. Its intervention resulted in a few product
demonstrations, but it was too little, too late. Geo-Info was abandoned and
its converted parcel data given to the collaborating county, which has found
it so valuable that it forms the basis of its county enterprise GIS system. The
data were also sold to cities that comprise that county. ES spent the next
four years successfully scaling down the project to a more practical level at
a fraction of the initial cost. An unfortunate event as the dying system was
nearing completion was that the utility company hired away almost half of
ES’s employees.
ES learned a lot from this early client failure and now spends more time
on making sure that there are business reasons for any new GIS system. ES
sees GIS development as different from IS development in the following
ways: (1) The IS unit performing ordinary IS development is unaware of the
visualization capability of their data and the benefits it can bring to services.
(2) GIS can serve as a centralizing force across the systems in a company. It
can reach across and serve to integrate a variety of isolated database systems,
displaying related information as easy-to-understand graphics (3) Different
levels of management regard the same spatial data through different lenses.
For example, upper-middle managers in companies tend to view data as a
way to analyze the business. Lower level managers view it as a technical tool,
while the CEO feels that all the areas are equally crucial. Spatial components
within the data allow a consolidated approach for users to view the same data
through different graphic perspectives.
ES also strives to design client spatial systems that see beyond the initial
limited system for the starting client. It believes that the consultant/outsourcer
needs to always keep one eye on the potential for broader applications across
the enterprise. Often users will get used to their first application success and
ask for more, or spread the word to others. ES wants to plan and make this
more possible by designing systems that can be scaled up and expanded.
Finally, a crucial part of consulting or outsourcing projects is the end stage,
i.e. how the consultant pulls away from the project and encourages the client
CHAPTER SUMMARY 193
to function independently. ES emphasizes training client-company support

and maintenance people, so ES can drop these and feel confident the system
will continue to function well. ES helps the client firm find and hire skilled
GIS personnel. By taking these steps, ES backs away from supporting GIS
and retreats to its original status as planning consultant to clients for future
projects.
ES has had dozens of successful projects for its governmental clients
since its founding and the early mishap with the utility client. Its better and
better capability to provide a strong systems development and implementation
framework for its clients is now one of its mission strengths. As represented
by JAD, ES tries to directly involve the key stakeholders and users at the
start in the planning of GIS projects, and continue to gain their “buy-off”
on the project as it progresses. The firm leverages its systems development
knowledge and technical skills to offset weaknesses in a particular client.
Since ES’s team works very closely with the client’s project team, even
working physically at the client premises, ES does run a risk of losing some
of its best field personnel, as happened early on with the utility firm and
continues to happen from time to time. ES not only illustrates the strength of
a formal systems development approach, but also demonstrates outsourcing,
from the outsourcer’s perspective.
CHAPTER SUMMARY
Systems development of GIS and spatial technologies is a crucial part of busi-

ness success with them. The reason is that an organization cannot just rest on
what GIS system it has now, but needs to continue to improve, update, and
expand. The chapter gives background on the approaches to systems devel-
opment, the phases of systems development projects, who does this, why the
users are crucial, and what some of the methods, techniques, and models are.
The cases in the book give some guidance about how GIS and spatial
applications are being developed and implemented in the real world. Based
on this limited sample, development has been mostly formal except for small
businesses. The approach is mostly traditional rather than object-oriented.
Development tends to be spearheaded by the GIS group. Outsourcing to out-
side firms has a fairly small presence in the sample, but two of the case
companies are outsourcers. The Engineering Systems case has provided in-
sight into the benefits of a formal, traditional approach, and highlights the
JAD technique, which early on brings together users, key stakeholders, and
the systems project team for rapid analysis and design. The case also gives
insight into the perspective of the outsourcer and recommends to businesses
that they maintain some of their own GIS capabilities, as the outsourcer can
only do so much, while some or all of the support eventually gets returned to
the client.
REFERENCES
Bowman, Scott. 2005. “Building an Enterprise GIS in a Limited Fiscal Environment.”

ArcUser, 8(1): 10–11.
Bowman, Scott. 2005. “Evaluating Enterprise GIS Requirements.” ArcUser, 8(1):
12–13.
Couger, J. Daniel, and Robert A. Zawacki. 1980. Motivating and Managing Computer
Personnel. New York: John Wiley and Sons.
2006 Proceedings.
ESRI 2000. “ModelBuilder for ArcView Spatial Analyst 2.” ESRI White Paper, May.
Redlands, CA: ESRI Inc.
Gray, P. 2006. Manager’s Guide to Making Decisions About Information Systems.
Gray, P., and H. Watson. 1998. Decision support in the data warehouse. Upper Saddle
Kearney, Pat. 2003. “Looking for Quality Geographic Data?” GeoWorld, 16(9):
38–40.
Malinowski, E., and E. Zimanyi. 2003. “Representing spatiality in a conceptual
multidimensional model.” In Proceedings of the 12th annual ACM international
workshop on geographic information systems, Washington, DC: ACM Press,
pp. 12–22.
Marble, Duane F. 2006. “Who Are We? Defining the Geospatial Workforce.” Geospa-
tial Solutions, May, pp. 14–21.
Spollen, Joseph F. 2006. “GIS Program Development: If At First You Don’t
Succeed—Try Again.” Proceedings of the Geospatial Information and Technology
Association, Conference 29, Aurora, CO: Geospatial Information and Technology
Association. Available at www.gita.org.
Valacich, Joseph S., Joey F. George, and Jeffrey A. Hoffer. 2004. Essentials of Systems
Analysis and Design, Second Edition. Upper Saddle River, NJ: Prentice Hall.
Viswanathan, N., J.B. Pick, W.J. Hettrick, and E. Ellsworth. 2005. “An analysis of
commonality in the twin metropolitan areas of San Diego, California and Tijuana,
Mexico.” Journal of Socio-Economic Planning Sciences, (39) 57–79.
West, Lawrence A. 2002. “Designing End-User Geographic Information Systems.” In
Tonya Barrier (ed.), Human Computer Interaction Development and Management,
CHAPTER 7
THE VALUE OF INVESTING IN GIS
GIS AND GEOGRAPHIC INFORMATION—THEIR VALUE
In considering GIS and spatial technologies, a business needs to ask the

question, are there net benefits coming from their use? That answer should
determine whether the company goes forward or not with GIS and spatial
applications. To assess the question, management needs to know what cate-
gories of costs and benefits are present, how to measure them, what methods
to employ in deciding whether there is net payoff, and what are the external
and intangible factors. The chapter considers how a cost-benefit analysis is
different for GIS and spatial systems compared to standard IT cost-benefit
analysis which has been studied more (King and Schrems, 1978; Devaraj and
Kohli, 2002).
The payoff from GIS is influenced by the effects of time. For example,
a medium-sized company invests a half million dollars over one year to re-
develop its client-server GIS into web-based service-oriented architecture and
implement it. When should the payoffs be measured? After a half year? Three
years? It’s even more difficult to compare the investments, mostly made early,
to the payoff if the value of the dollar is eroding over time through inflation.
The technique of Net Present Value (NPV) is presented that allows time and
inflation factors to be taken into account. The chapter discusses the steps
a firm can take to conduct cost-benefit analysis. The books’ cases demon-
strate that large firms tend to utilize more formal methods for determining
the payoffs from GIS, while small ones usually do it informally, sometimes
by “seat of the pants.” Some studies have shown that IT investments do
195
196 THE VALUE OF INVESTING IN GIS
not yield net measurable benefits. This issue, referred to as the “productiv-
ity paradox” (Brynjolfsson, 1993; Lucas, 1999), is taken up regarding GIS
investments.
Case studies of Sears and Norwich Union are considered. Sears developed
extensive methods for cost-benefit analysis, while the Norwich Union case
demonstrates how unexpected benefits can occur from discovering the value
of spatial data.
Why perform cost-benefit analysis for GIS? Three of the important reasons
are as follows (King and Schrems, 1978; Worrall, 1994):
1. Cost-benefit (C-B) analysis assists in planning for an organization. Plan-

ning involves tradeoffs between competing demand for organizational
investment and resources. C-B analysis can help in deciding between
competing demands. GIS becomes one of the competing demands and
must rise high enough in planning priority to compete.
2. Cost-benefit analysis is useful in auditing. The organization may decide
after a GIS system has been put into effect to perform a retrospective
C-B analysis, to audit what occurred. The auditor determines whether
or not the investment paid off, and suggests changes.
3. To prepare and support participants in political decision-making. This
is less formal and more rapid than the planning uses in 1 and 2. It is
done with less complete information, but is more commonplace than
the other reasons (King and Schrems, 1978). Often budget decisions
are partisan and depend on the power structure of the players involved.
However, in that environment the parties who know the C-B details may
do better.
SETTING THE CONTEXT FOR COST-BENEFIT ANALYSIS
The stage and context needs to be set at the start to do a C-B analysis. As a
C-B project is launched, the following four factors are useful to clarify early
on (King and Schrems, 1978):
1. Statement of purpose. This statement indicates whether the C-B analysis

is being utilized directly to make a decision, to provide background data
for a decision, or to politically influence decision-making. It is important
to ascertain, since the methods, quality of the data, and the reporting of
findings differ by purpose.
2. Time simultaneity. The C-B analysis must indicate whether the C-B
analysis pertains to a future GIS system, involves a current GIS system,
or is retrospective. Any time point may be useful for C-B.
3. Scope. The C-B analysis may be comprehensive in examining all pos-
sible costs and benefits. On the other hand, it may be severely limited
SETTING THE CONTEXT FOR COST-BENEFIT ANALYSIS 197
in scope, for instance, it may only manipulate a single cost item, as part
of a sensitivity analysis, to see how it affects the benefit outcomes.
4. Criterion. The last contextual factor is the method that is used to com-
pare the costs and benefits, after they have been calculated. A problem
is that costs and especially benefits may not be measurable in monetary
terms. For instance one benefit of a new spatial web site is to improve
the company’s brand image. However, the analyst is not able to convert
this to monetary values. If there are intangibles, they may or may not
factor into the criterion. Intangible is defined as not convertible into a
currency value.
The criterion may be a quantitative index measure, graphical compar-
ison, or involve a lengthier model. The criterion definition needs also to
state whether or not the values of the costs and benefits are to take into
account inflation and opportunity costs through net-present-value cal-
culations. Also, the criterion definition needs to indicate if intangibles
are excluded or included.
Among the challenges in GIS cost-benefit analysis is that GIS tends to have
higher costs than for conventional information systems, due to its considerable
data acquisition and data management needs. This is because a GIS is based
on both attribute data and spatial data (Huxhold, 1991; Clarke, 2001). The
extra time and effort, versus non-spatial IT, stems from the need to do the
following:
Gather boundary data and associated attributes

Convert the data to digital form
Design or configure topological data structures and link non-spatial with
spatial data
Maintain GIS boundary files and data
Estimates indicate data collection for GIS may constitute 65 to 80 percent

of the total cost for systems development and implementation (Huxhold and
Levinsohn, 1995; Obermeyer, 1999; Tomlinson, 2003). Further, the attribute
and digital boundary data need to be linked together. These linkage tasks
add to costs, relative to non-spatial IS. Another difference has to do with
GIS’s feature of visualization (Jarupathirun and Zahedi, 2004). In this re-
spect, GIS is comparable to a multimedia business application, which may
have higher costs, due to its visual aspects. GIS tends to have a higher pro-
portion of intangible benefits than a non-spatial IS. In some cases, it may
be possible to build models that convert the benefits into a tangible ben-
efit, but the organization does not commit the money to accomplish the
conversion.
Another difference is that GIS applications often depend on outside data
providers who create risk in two ways: (1) lack of commitment to an ongoing
Figure 7.1 Comparison of Cost-Benefit for a Simple GIS System to a GIS System
Closely Coupled with Another System or Technology
cycle of data import, conversion to proper format, and linkages within and
between complex spatial datasets; (2) external data supply may be interrupted
which might render the GIS system worthless. For instance, a chain of hos-
pitals models and maps the traffic patterns that affect patients’ access. The
highway data are provided by a state university consortium. However, the
consortium loses its funding, leading to shut-down of the data flow. IT also
has these problems, but the cycles of data refresh are usually less complex
and often there are more options if one data source shuts down.
Another distinctive aspect of GIS from Chapter 1 is that GIS is frequently
coupled with other software systems and technologies. The systems and
technologies such as GPS with which GIS is coupled appear in Table 1.3. As
seen in Figure 7.1, because of this coupling, the cost and benefit calculation
for any one of them may be more difficult.
Figure 7.1 demonstrates that the assessment of costs and benefits of an un-
coupled GIS system can be done by performing a straightforward assessment
of that system’s costs and benefits, followed by a cost-benefit comparison
with a criterion such as break-even analysis. However, the cost-benefit anal-
ysis becomes more complicated for a GIS coupled with another system or
THE COSTS AND BENEFITS OF GIS 199
technology such as handheld computer, RFID, or GPS. For multi-component,

linked systems, it is difficult to disaggregate the costs and benefits of the GIS
from those of the other technologies. Additionally, the two coupled compo-
nents may be at different stages of development, which complicates NPV and
break-even analysis.
For a multi-component system, the marginal cost of investing in one spatial
component to add to an existing system may be at a much lower cost than the
total cost of implementing the full system from scratch. This must be weighed
against the level of benefits for the two alternatives.
THE COSTS AND BENEFITS OF GIS
This section discusses the categories of costs and benefits of a geographic

information system. Costs and benefits may be divided into tangible i.e. able
to be converted into monetary amounts and intangible i.e. not convertible
to monetary values (King and Schrems, 1978; Obermeyer, 1999; Karikari
and Stillwell, 2005). Costs for information systems including GIS are pre-
dominantly tangible, while benefits are a mixture of tangible and intangible.
For instance, the cost of a GIS managerial employee can be estimated by the
tangible value of his/her salary and job benefits. However, the most significant
benefit of the employee is his/her effective leadership and decision-making,
outcomes difficult if not impossible to convert into dollar amounts.
Costs
The costs of a GIS can be classified into the categories given in Table 7.1.
The costs are all tangible and possible to estimate. (Note: intangible costs are
not considered here.) However, some problems often occur in calculating the
costs, and there is a risk of excluding others that might be relevant. Many
common errors prevalent in cost accounting create pitfalls (King and Schrems,
1978), such as not identifying hidden costs, counting costs twice, or omitting
important costs. An example of hidden costs would be costs located at other
places in the organization that are not being counted against the GIS. For
instance, the development of a GIS system depends on the ideas of the GIS
team leader, who spends a lot of time in discussion with analysts and managers
in Central IT. However, the overhead of the time spent by the GIS team leader
with Central IT is not included in the costs. Omitted costs are ones that are
not obvious, but are in fact dedicated to the project. For example, space, site,
and utility costs are commonly omitted, but may be important particularly in
expensive locations, such as downtown Chicago.
Risk management has tangible and intangible costs. Among the tangible
costs are security and insurance. Security reduces risk and insurance invests
in mitigating its impacts. There are also intangible costs associated with risks.
If a customer-centric GIS fails, such as a web service that provides routing
TABLE 7.1 Tangible Costs of a GIS
Hardware
Software
Data collection (internal)
Collection or purchase of external data
Staff
Transformation of manual maps and data into digital format
Systems development and implementation Business process change
Project management
Web services
Maintenance costs for hardware and software
Maintenance of data
Supplies
Training
Outsourcing (e.g. GIS applications programming)
Consulting
Licensing
Security
Insurance
The internet, other networks, communications interfaces
Space, site, and utilities
Modified and expanded from Huxhold, 1991; Worrall, 1992; Obermeyer, 1999;
and Tomlinson, 2003
to customers, corporate image and reputation may be muddied but the costs
can’t be quantified. Other examples of risks for spatial systems that may
incur intangible costs are privacy intrusion, reduced safety (e.g. if criminal
targets appears on public maps), user resistance, and added intelligence of
competitors through public maps. In many circumstances, there is a low
probability of cost, but it could be substantial.
Benefits
The benefits of GIS and spatial technologies are more difficult to measure than
costs (King and Schrems, 1978; Obermeyer, 2000; Worrall, 1994; Tomlinson,
2003; Karikari and Stillwell, 2005; Maguire et al., 2007). The reason is
that benefits are expressed as an informed, ready, efficient, and decisive
organization, which is hard to measure, since there are many beneficiaries,
time lags, and intervening causes (Table 7.2). Benefits may also be expressed
in terms of greater information value, but the value of information is difficult
to measure because it depends on timeliness, its users, and what decisions
it is influencing (Ahituv, 1989; Ragowsky et al., 2000). Another problem
with measuring benefits has to do with quantity versus quality. How do you
THE COSTS AND BENEFITS OF GIS 201
TABLE 7.2 Benefits of GIS (Tangible and Intangible)
Tangible
Total salary and benefits lowered through reducing the workforce
Improved productivity
Improved performance
Greater accuracy
Higher value of assets
Cost reduction (through employees performing their tasks more efficiently)
Cost avoidance in the future (projected greater workload per employee)
Time saving in searching, manipulating, and processing maps and data.
Expansion of revenues (achieved through improved data quality, speed, efficiency)
Intangible
Improved decision making
Effectiveness of managers and executives
Reaching strategic objectives
Environmental scanning
Visualizing complex data
Improved siting of offices, branches, factories, and other facilities
Timeliness of information and quicker response
Volume and quality of information
Better capability to sell products (by spatial web services, manual maps)
Improved customer service
Identification of missing revenue sources (e.g. in government, identification of properties
not being taxed)
Better environmental scanning
Reduced error
Reduced liability (e.g. GIS for security monitoring)
Improved brand image
Modified and expanded from Huxhold, 1991; Obermeyer, 1999; Stein and Nasib, 1997; Devaraj and Kohli,
2002; Tomlinson, 2003; and Karikari and Stillwell, 2005.
measure the value of a GIS that produces fewer, high-resolution maps versus
one producing numerous low-resolution ones?
Although more challenging to measure than costs, benefits can be calcu-
lated. Suggestions on ways to compute them are the following:
1. Disaggregate larger benefits into smaller pieces, which may be more

amenable to quantification. Consider an example. An advertising firm
has identified the benefit of better knowledge of its products through
presentations to its customers that rely on GIS. However, the overall
better knowledge cannot be made tangible. Nevertheless, if the presen-
tation process is broken down into a set of small control items, such as
number of maps, length of the average presentation, number of sales
agents making presentations, the benefits of GIS may be more amenable
to assignment of dollar values, although it is still not easy.
2. The C-B analysis can be restricted to only tangible costs and tangible
benefits (Smith and Tomlinson, 1992; Obermeyer, 1999). If that re-
sult indicates net benefit, then the intangible benefits may be regarded
as an added plus. Another variation on this approach is to perform a
break-even analysis for a future time point, just restricted to tangibles
(King and Schrems, 1978). If costs exceed benefits, the negative value
of benefits minus costs can be compared against positive intangible
benefits. They may be close enough to yield a compelling argument for
implementation.
An example of benefits calculation is for Lamar Advertising Company,

which is the second largest billboard company in the United States. Lamar
depends on nearly 200 sales managers, who sell billboard contracts to ad-
vertising companies and their client firms. The managers benefit by well
documented reports that include maps indicating potential billboard locations
relative to traffic flows and other features. The benefits of GIS are mainly
improvement in closing contracts, and also improved marketing planning.
Tangible benefits can be estimated by seeking expert opinion to gauge the
percentage improvement in contract closings from introducing GIS. This
change in percentage can also be measured since closing percentages were
measured before GIS was introduced and afterwards. The percentage im-
provement can be multiplied by annual sales to determine the dollar benefit
annually. Finally, intangible benefits include a better corporate image for
Lamar and better responsiveness of sales managers.
Benefits calculations can be helped in some cases by creating simulation
models of corporate revenue-generating or profit-generating activities that
measure the sensitivity of adding GIS. The difference before and after is an
estimate of tangible benefit.
PREVALENCE OF COSTS AND BENEFITS IN CASE STUDY

COMPANIES
The book’s research case study companies indicate overall patterns in cost-
benefit analysis. Although a small sample, it gives some insight into overall
industry trends in spatial costs and benefits. Two-thirds of the companies
indicated that the level of net GIS benefits, i.e. benefits minus costs, was high.
This supports a finding that for these companies GIS and spatial technologies
are paying off. Only one firm, Large Credit Bank (LCB) indicated low net
GIS benefits. LCB is a very large bank with a monoline product line in the
credit card area, but has acquired several medium-sized regional consumer
banks. IT in the credit bank is dominated by legacy systems. Although the
acquired banks are users of GIS for branch siting, the benefits are restrained
due to a top management in the big central unit that doesn’t understand GIS
and is resistant to it.
PREVALENCE OF COSTS AND BENEFITS IN CASE STUDY COMPANIES 203
Only the largest firms, Sears, The Southern Company, GIO, URS, and
Large Insurance Company, used formal methods of cost-benefit analysis. The
methods were ROI and Net Present Value (NPV). The GIS managers in the
large firms pointed out that these methods were company-wide standards
applying to all corporate projects, and were not chosen specifically for GIS.
The Southern Company did thorough NPV analysis broken out for each
major cost category, for every operating company, and for the Southern-
wide aggregate. This level of detail is helpful to upper-middle management
in budgeting for, and tracking Southern’s ambitious enterprise-wide spatial
initiative. The largest firms, GIO and Sears, singled out very large GIS projects
as having more rigorous cost-benefit models and complex financial analysis.
It is firms of such size that would tend to have financial staff available to
conduct such studies.
Across all firms, there was good agreement on the cost categories of per-
sonnel, software, hardware, and data (see Table 7.3), all of which appear in
Table 7.1. The near agreement on costs to be considered probably stems from
their more tangible nature. Furthermore, the components needed to develop
and build spatial applications are rather uniform, even though the applications
are varied. A number of firms such as Large Insurance Company (LIC) and
Norwich Union, cited data costs as being their highest. Both insurance firms
purchased a lot of data from external sources, so those costs were high and
also visible. Firms gave less emphasis to the cost of internal data. It is likely
internal data costs are also high, but sometimes not broken out and not recog-
nized in internal budgets or possibly the GIS use was incremental to existing
uses of the data and not so significant.
No firm mentioned outsourcing as a cost item, pointing to its lesser presence
for GIS versus its rampant role in more conventional systems. Its reduced
extent may be due to: (1) GIS’s specialized nature, (2) its importance in
some firms as a core competency, obviating emphasis on outsourcing, and (3)
shortages of spatially skilled people in the outsourcers. The topic is returned
to in Chapter 10.
Benefits of GIS for the case firms, shown in Table 7.4, are quite varied,
which corresponds to their less tangible nature, the broad variety of benefits
already mentioned (see Table 7.2), and the wide differences expected for
firms with different products, industries, customers, and strategies. Among
benefits more prevalent for the case companies are improved facilities siting,
quicker response, and better decision-making. These are fundamental benefits
that managers should look for when implementing a GIS. Several companies
pointed to the benefits derived from the value of certain spatial data, some so
valuable it might be marketed as a separate product. One such firm, Norwich
Union, will be discussed as a chapter case later. Four-fifths of the benefits cited
are intangible, confirming tangible costs, but intangible benefits. However, it’s
important to keep in mind that “intangible” sometimes means that there are
not the tools available nor the management commitment to divert sufficient
company resources to measure it tangibly.
c07
TABLE 7.3 Costs of GIS for Case Sample of Firms
Uses C/B
Formal Level of Net Strategic
Name or Description Size Method Method Used GIS Benefits Level Costs
Global Integrated Oil large Yes ROI and others for large high very high people, data (largest cost),
(description) projects hardware, software
URS large Yes Estimate simple high medium people, software, hardware
tradeoffs.
Large Credit Bank large No low low people, data, software, hardware
(description)
October 18, 2007
Rand McNally large No high high people, data, software tools,

opportunity costs
14:25
Southern Company large Yes NPV for each cost high high hardware, software, IT support,
204
category, aggregated other support, personnel
Southern-wide and for
each operating company.
Sears Roebuck large Yes NPV. Compute paybasck high high people, data, hardware, software
Char Count=
periods. For large scale development

projects, more detailed,
rigorous financial
models are used.
Kaiser Permanente large No medium medium people, hardware, software,
outsourcing
Large Insurance large Yes ROI high high salary and benefits of GIS team,
Company hardware, software, large data cost
(description) for external data.
c07
Large Commercial large No high low software, hardware (desktops and

Bank (description) printer) salary and benefits,
training, data.
Norwich Union large No high high software, hardware, data (a lot

external), personnel.
Baystate Health medium No medium low undetermined
Lamar Advertising medium No high high people, hardware, software, web
Company services
October 18, 2007
Arizona Republic medium No medium high people, hardware, software, data

Chico’s medium No medium high data (most expensive and from
outside), software, people.
14:25
Hardware is in IT budget.
205
Western Exterminator medium No medium medium people, hardware, software
Sperry Van Ness medium No high high personnel, hardware, software -
Microsoft, training package
Prudential Preferred medium No medium high software, hardware, personnel. All
Char Count=
Realty is shared with IT and other

functions.
Engineering Systems small No high low people, hardware, software
MapGistics small No high high training, hardware, software,
financial costs
Motion-Based small No high very high people, web services
Technologies
ROI = return on investment. NPV = net present value.

TABLE 7.4 Benefits of GIS Case Sample of Firms
Name or description Size Benefits
Global Integrated Oil large Know where assets located. Track assets over time.
(description)
URS large Better quality of output. Efficiency. The value of the data
in the maps.
Large Credit Bank large Analyze new branches and facilities.
(description)
Rand McNally large Better decisions on marketing Rand McNally products.
Make high quality map products.
Southern Company large Consolidation benefits from GIS enterprise-wide.
Identify substandard geographies for cost avoidance.
Sears Roebuck large Increased productivity for a huge workforce. Faster
response. Stretch out periods of GIS capital investments.
Kaiser Permanente large Decision-making, better government submissions,
presentations, identification of catchments areas.
Large Insurance large Increase underwriter profit. Write better policies.
Company (description) Decision-making.
Large Commercial large Added revenue from making good branch location
Bank (description) decisions. Support to sales campaigns that bring in
customer dollars and business relationships.
Norwich Union Large Revenues from GIS for floods, better profits from
customer base with GPS in cars. New data products.
Baystate Health medium Undetermined, since GIS projects are recent.
Lamar Advertising medium Closing contracts (maps are a key element in proposals
Company to advertisers). Better control of marketing development
expenses. Mapping available to sales team at 160 sites.
Arizona Republic medium Competitive edge.
Chico’s medium Forecasting volume of merchandise and related sales
and staffing. Accurate forecasting benefits suppliers.
Improved marketing and response time to supply stores.
Western Exterminator medium Improved decision-making for marketing. Better siting
of new centers. Better routing.
Sperry Van Ness medium Savings in employee time. Speed to market in making
proposals to clients.
Prudential Preferred medium Pleasing customers with new spatial web functions and
Realty tools leading to better customer retention. Better
locational assessments by executive team of listings and
agent performance
Engineering Systems small Revenues from clients for GIS projects that the firm
completed.
MapGistics small Improved bed occupancy efficiency, speed of patient
discharge, locating assets.
Motion-Based small Increase in proportion of web customers with paid
Technologies subscriptions.
PREVALENCE OF COSTS AND BENEFITS IN CASE STUDY COMPANIES 207
Based on the limited case sample, we can conclude that (1) GIS is bene-
ficial, (2) informal “seat of the pants” methods are mostly used to measure
the net benefits, except for very large firms, and (3) while costs are tangible,
benefits are a mixture of tangible and intangible.
Benefits vary a lot by firm and industry circumstance. For example, since
GIO as an exploration and production company needs to know where its
huge base of assets is located over time, GIS provides its greatest benefits
in asset tracking. The benefits are intangible. Kaiser Permanente has varied
GIS applications, so its benefits are also diverse and mostly intangible. Large
Insurance Company focuses its greatest use of GIS on underwriting and
policies and that is where the mixed tangible-intangible benefits are. The
tangible benefits for Sperry Van Ness are speed of response and better use
of time for its national marketing managers. That’s where its GIS is focused.
In nearly all the cases, the benefits are specialized and keyed to the specific
focus on the GIS applications, rather than generic ones. This reflects that GIS
has not yet risen to become a tool in general use across enterprises.
Unit of Analysis
The difficulty in measuring the costs and benefits of GIS varies by the unit
of analysis (King and Schrems, 1978; Obermeyer, 1999; Tomlinson, 2003).
Among the units of analysis for GIS are the following:
Industry
Company
Department or division
Project
Individual
Considering the different units is useful to organizations performing

C-B analysis, as well as to industry analysts and researchers studying trends
and doing comparisons. C-B analysis is mostly commonly performed at the
company level, and is easier to do where GIS is smaller but challenging if it’s
enterprise-wide for a larger firm. For budget analysts, it can be beneficial to
disaggregate costs and benefits below the company level. Firms do not usually
perform C-B analysis for GIS at the individual level as data are missing or
it’s too expensive, although it might apply for a small business.
Analyzing C-B at the project level offers a number of advantages. Firms
need well-controlled project management procedures to track the project costs
and benefits over time, with regular reporting intervals. This tracking can then
be input to accomplish ROI and NPV calculations. Having the project detail
helps a firm in its GIS planning, coordination, and auditing. For instance,
benefits for GIS Project A are realized evenly for each of the five years after
implementation, versus for GIS Project B, benefits are realized on a declining
basis – 40 percent in the first year, 30 percent in year 2, 20 percent in year
3, 10 percent in year 4 and none in year 5. Having the project detail allows
management to better plan its annual cash flows, and make adjustments to the
projects if necessary.
Similar advantages of disaggregation apply by departments and divisions.
The disadvantage to disaggregating is the higher data collection, monitoring
and coordination costs of the C-B analysis. C-B analysis at the industry level
is usually done by consultants or researchers.
For projects with small scope, such as a single-user desktop project, costs
and benefits may be more readily estimated. For them, there is limited inte-
gration with other systems and technologies, intervening factors are reduced,
and the external environment is not as influential. On the other hand, if the
unit is a corporate-wide GIS system, it may be challenging to separate its
costs and benefits from those of other systems inside the company, from
inter-organizational systems, or from the outside environment. Furthermore,
for enterprise-wide GIS, the attribute data are commonly shared with other
company systems, such as marketing, or with the firm’s enterprise resource
planning (ERP) system. Inter-organizational GIS systems are even more chal-
lenging to assess. The shared aspect of the business data complicates the
separation of GIS’s costs and benefits. Since the trend in GIS is towards
enterprise-wide web-based systems, the challenge of determining costs, ben-
efits, and payoff is likely to grow.
THE SUBSTEPS IN COST-BENEFIT ANALYSIS FROM SYSTEMS

DESIGN PERSPECTIVE
The last chapter examined the many dimensions of systems analysis and
design for GIS. This section briefly considers the project steps in the C-
B analysis, which is part of the feasibility stage in systems analysis. It is
intended to encourage managers to plan the steps of C-B analysis carefully.
The following are the standard cost-benefit substeps for IT:
1. Develop an overall plan for the C-B analysis to be conducted over the
system useful life.
2. Decide on the analyst or analyst team.
3. Consult with executive management and key user managers to identify
which are the largest and most important categories of costs and benefits.
4. Determine the alternative C-B analyses to be performed . Decide on the
unit of analysis, company, or by departments, or projects.
5. Determine all the material factors for costs and all the material factors
for benefits.
6. For each tangible factor, decide how it will be measured.
7. Decide if intangible costs and benefits will be included. Decide on how
they will be evaluated.
THE SUBSTEPS IN COST-BENEFIT ANALYSIS FROM SYSTEMS DESIGN PERSPECTIVE 209
8. Measure the tangible costs and benefits. For benefits, the measurement
methods include expert opinions, simulation modeling, and experimen-
tal comparisons.
9. If included, evaluate the intangible costs and benefits.
10. Compare the cost-benefit results over the entire time period of the study.
Include financial summary measures such as break-even point, etc. and
criteria to evaluate alternatives. If appropriate, conduct discounted cash
flow analysis.
11. Perform a comparative analysis of the alternatives from step 4.
12. Decide on what recommendations to make to management, based on
these results. Present the findings to management.
(Modified from King and Schrems, 1978; Ahituv and Neumann,
1990; Karikari and Stillwell, 2005; Maguire et al., 2007)
Before commencing C-B analysis, an analyst must be selected—a choice

that is often critical for C-B’s success or failure. There are a number of
places to find the analyst (Ahituv and Neumann, 1990): (1) Inside person.
This is someone within the organization who has the necessary financial
management and technical skills, as well as experience and knowledge in GIS.
(2) Outside consultant. Auditing, accounting, and GIS and IS consulting firms
can provide such persons. (3) Persons from other organization(s). Someone
from an affiliated organization may be loaned or made responsible to perform
the C-B analysis. They may stem from government or corporate oversight, or
through corporate alliances.
For GIS, the C-B substep steps differ in some respects. For Step 2 (develop
an analyst team), GIS is influenced by its historical origin in the public sector,
outside of the mainstream of business IT (Huxhold, 1991; Tomlinson, 2003).
Thus, analysts from the public sector may be less well-trained in formal
systems analysis and C-B techniques than those coming from careers in
business or information systems. For Step 5 (determine the material factors
for costs and benefits), the material benefits may be reduced or more difficult
to measure, due to the presence of visualization. In Step 10 (comparative
analysis), comparisons may be complicated by the problem discussed earlier
of close coupling between GIS and other spatial systems and technologies.
How does GIS differ from non-spatial IS in relative importance of different
categories of costs and benefits, given in Tables 7.1 and 7.2. For GIS, these
categories are influenced by the relatively greater costs for acquiring spatial
data (Huxhold and Levinsohn, 1995; Obermeyer, 1999; Tomlinson, 2003).
The capital costs of data acquisition vary by whether the data and coverages
come from the public or private domains. Since historically GISs were mainly
in the public domain, large banks of public data and boundary files have been
available for free or very low cost in the U.S. An example is the U.S. Cen-
sus maps and associated data on population, housing, social characteristics,
and the economy, distributed free over the internet (www.census.gov) (U.S.
Census Bureau, 2007). Other free geospatial data are available U.S. pub-
lic portals such as GeoData.gov (www.geodata.gov), and NationalAtlas.gov
(www.nationalatlas.gov). Internationally data are available from such portals
as The Geography Network (www.thegeographynetwork.com), and Go-Geo!
for the UK (www.gogeo.uc.uk).
The time spent organizing the data, entering it into databases such as Oracle,
and checking it for quality, sometimes makes up for the “free availability”
from governments. Another source of data is third-party service firms that
make government and private sector data available in better organized form,
but at cost. On the other hand, business proprietary data can be quite costly or
unavailable for competitive reasons. In Europe and most of the world outside
the U.S., the cost of government spatial data is often high, which adds to the
data cost problem.
COMPARISON OF COSTS AND BENEFITS OVER TIME
As was seen in Figure 7.1, the concluding step in a cost-benefit analysis is

to compare the costs against the benefits in order to determine whether or
not the GIS investment has net benefit or not. Graphical and non-graphical
methods are available to perform comparative analysis of cost-benefit results
over time (King and Schrems, 1978; Kingma, 2001; Boardman, Greenberg,
Vining, and Weimer, 2001; Karikari and Stillwell, 2005). Among the more
prominent ones are the following.
r Break even point
◦ The total investment cost is divided by the annual benefit of the GIS.
This gives the number of years to break-even. However, this calculation
may need to be adjusted for a multi-year investment due to changes in
the time value of money.
r Formulae
◦ Costs and benefits are directly compared through a formula.
r Baseline cost comparison charts
◦ Use of a graph to compare the annual cost of running an enterprise
without GIS and the cost of running the enterprise with GIS. The charts
can show GIS costs savings annually or cumulatively, and can plot net
present value on tangible benefits at different discount rate assumptions.
A benefits yield curve indicates the percent of cumulative benefits
realized from a GIS project or from GIS in the firm over a multi-year
period (Maguire et al., 2007).
r Discounted cash flow
◦ A common technique of cost-benefit analysis is to account for the
discounted value of money over time. Both costs and benefits are dis-
counted over time, assuming a regime of future discount rates for infla-
tion. This is accomplished by the calculation of net present value.
COMPARISON WITH A CRITERION 211
THE METHOD OF DISCOUNTED CASH FLOW
The net present value (NPV) for a cost or benefit for a particular year in the
future may be estimated by:

n
xt
NPV =
t=0
(1 + d)t
where NPV is the present value, xt is the cost or benefit value during time
period t, and d is the discount rate. After choosing appropriate discount rate
or rates, all the future costs and benefits can be estimated.
The present value of the costs and benefits can be summarized together in
the equation:
n
Bt − Ct
NPV =
t=0
(1 + d)t
where Bt represents the benefit at time t and Ct represents the cost at time t.
There are fine points in calculating the net present value that are beyond
the chapter’s scope, but can be found in books on cost-benefit and financial
accounting. They include at what point in a time period the inflation rate
is calculated (beginning, middle, end); whether the inflation rate fluctuates
(this equation assumes it remains steady); and whether costs are influenced
differently by inflation than benefits (King and Schrems, 1978). The analyst
needs to take into account that the precision of the cost and benefit figures
will influence the accuracy of this calculation.
Since GIS projects tend to have long periods from start to break-even due
to the higher investment in start-up and data acquisition (Tomlinson, 2003),
the impact of net present value calculations may be large.
COMPARISON WITH A CRITERION
Usually the C-B analysis involves several alternatives having different values
for costs and benefits. A criterion is used to compare them. Let’s say that an
analyst has prepared six alternatives under different cost-benefit assumptions.
A criterion is used to select the most suitable one (King and Schrems, 1978).
Among the criteria commonly used are:
r Maximize the present value of the benefits minus the present value of the
costs.
r Maximize the ratio of benefits over costs.
r Assume a given level of costs for all alternatives and maximize benefits.
r Assume a given level of benefits for all alternatives and minimize costs.
Any of these criteria, or others, can be used. The choice depends on the
particular problem and context of management decision-making.
A problem can occur if insufficient alternatives are examined for the prob-
lem at hand. For instance, for a small-scale GIS with a single data source,
single boundary file, and clear uses, perhaps two or three alternatives would
be appropriate. By contrast, for an enterprise-wide, web-based GIS involv-
ing millions of dollars in expenditure and tens of thousands of users, more
alternatives are needed. Even for a large scale, complex GIS, however, it is
not possible to include all the alternatives. Experience has shown that results
may be improved, if the key interested parties are involved in determin-
ing which alternatives to include (King and Schrems, 1978; Maguire et al.,
2007).
INTANGIBLE COSTS AND BENEFITS
Intangible costs and benefits are prevalent in GIS applications (Obermeyer,

1999; Tomlinson, 2003; Karikari and Stillwell, 2005). Some examples of
intangible benefits are the following (Ahituv and Neumann, 1990; Tomlinson,
2003; Karikari and Stillwell, 2005):
r Image improvement of the organization

◦ Better planning
r Better decision-making
◦ Improved customer service
◦ Better understanding through visualization
r Enhanced employee morale
r Improved information to executives
After determining the intangible costs and benefits, the intangible costs and
benefits can be prioritized by interviewing managers in user departments The
intangible C-Bs are presented in summary form to management, along with
the tangible ones (Ahituv and Neumann, 1990; Karikari and Stillwell, 2005).
Some experts advise, in meeting with top management, to justify the net
tangible benefits first, and then present the intangibles (Maguire et al., 2007).
No matter how the C-Bs are presented, top management can ultimately decide
how much to weigh the intangibles.
What is distinctive for GIS versus non-spatial IS concerning intangibles?
First, the visualization capabilities of GIS may provide improved decision
quality for certain classes of decisions (Mennecke et al., 2000; Swink and
Speier, 2000), but the degree of improvement may be difficult to quantify. Vi-
sual responses tend to be difficult to measure. For example, if there are two GIS
systems, where the first one produces low resolution maps and the second one
yields maps with five-fold better resolution, how can the advantage of this
STAKEHOLDERS AND EXTERNALITIES 213
intangible benefit on business effectiveness and decision-making be mea-

sured? A second feature of GIS that has stimulated more intangibles is the
tendency of GIS to move up in the hierarchy of business applications to be-
come more strategic once the benefits are realized. At the strategic/competitive
level, the benefits are less tangible versus at the operational or middle man-
agement levels (Ahituv, 1989).
THE FEASIBILITY DECISION
Management must eventually weigh the results of a cost-benefit analysis and

make a decision on one of the alternatives, including staying with the sta-
tus quo (Ahituv and Neumann, 1990; Satzinger, Jackson, and Burd, 2002).
Analysis of feasibility for IT systems is divided into three areas: (1) financial
feasibility, (2) technical feasibility, and (3) institutional feasibility. Feasibility
decisions are influenced by the total time period of commitment before the
decision is revisited, for instance one year, five years, or ten years. If the fea-
sibility decision needs to hold for ten years, then a much more in-depth study
must be undertaken. No matter what depth, as the time frame extends out,
it becomes increasingly difficult to accurately predict technological changes
(Day and Schoemaker, 2000).
For financial feasibility, GIS is similar to a conventional IT application. It
is based on the cost-benefit comparison described in the last section. Deter-
mination of technical feasibility for GIS follows standard methods detailed
elsewhere (Satzinger et al., 2002), but may put more emphasis on the feasibil-
ity of linking GIS with associated technologies such as GPS, RRID, sensors,
and mobile services. Institutional feasibility refers to the capability of the
institution to support a GIS project. GIS requires specialized human workers,
with sufficient knowledge to carry out a GIS project, either inside the institu-
tion or present in an outsourcer. Institutional feasibility is affected by GIS’s
distinctive features, especially visualization capability and presence of linked
systems.
STAKEHOLDERS AND EXTERNALITIES
Cost-benefit outcomes vary depending on the vantage points of different

stakeholders to an organization. For instance, GIS in an advertising firm will
have different costs and benefits for the corporation itself, marketing and
advertising people in client firms, investors, and the public who eventually
view the advertisements. The effort of cost-benefit work is made more diffi-
cult and time-consuming by conducting separate C-B analysis for different
stakeholder groups (Obermeyer, 1999). However, it may be worth the effort,
if there is wide diversity in the stakeholder perspectives, and if stakeholders
are critical elements in firm success. At the minimum, stakeholders should be

mentioned in the cost-benefit analysis.
Positive and negative externalities of a GIS refer to the indirect impacts
of system implementation. This can be seen by the analogy of environmental
externalities of an industry process. A process intended for manufacturing a
product may cause the indirect effects of pollution, noise, and human injuries.
An example of a negative externality for GIS is the loss of the information
security or competitive advantage in an environmental company, as its GIS
system for environmental mitigation becomes more widely deployed. By
contrast, positive externalities may arise from Google mashups that provide
free GIS and mapping to the public, or from the GIS-driven website of a local
government that provides GIS analysis and mapping to the public. There may
be unintended benefits, for instance high school students visiting the mashup
websites and learning about products and services might be converted as
customers in a few years.
CASE STUDY OF GIS COSTS AND BENEFITS: SEARS ROEBUCK
The giant retailer has extensive enterprise spatial technologies in six areas:
routing and deploying service technicians, routing/deploying home delivery
drivers, warehouse optimization, demand forecasting in marketing, automated
vehicle navigation, and capacity management of the technician workforce in
the service territories (Jones, 2005). The scale is huge, for instance this
set of tools, referred to as the Sears Smart Toolbox (SST) supports 10,000
technicians in all fifty states, and its GIS/GPS systems support eleven million
in-home service orders yearly (Jones, 2005). The systems have been deployed
carefully and cost-efficiently. There have been major gains in productivity of
delivery and service maintenance that have enhanced the company’s bottom
line. As the leading firm using spatial technologies in its huge markets, there
is strong competitive advantage.
The six coordinated home product repair and delivery system modules
are shown in Figure 7.2. The in-vehicle navigation is shown at the lower
right. The support systems for home deliveries are the three on the upper
part of the diagram: Demand Forecasting in the upper left, Enhanced Home
Delivery System to determine the optimal home delivery route in the upper
center, and the Warehouse Optimization with GIS in the upper right. The
latter system does spatial routing for forklifts in the warehouses to pick the
merchandise for each truck loading. The CAMS and CARS systems (lower
left and lower center) assess the capacity of service technicians in geographic
areas (CAMS) and determine the optimal routing for service technicians
(CARS). Mobile Mapping (lower right) supports naviagation of the truck
drivers. All six systems in the Sears Smart Toolbox draw data from the same
mainframe system, and the six systems communicate with each other through
radio links and satellite.
CASE STUDY OF GIS COSTS AND BENEFITS: SEARS ROEBUCK 215
Figure 7.2 Sears Integrated GIS Systems for Product Repair Services and Home
Delivery. Source: Jones, 2005
Sears has invested significantly in long-term and large-scale GIS systems.

Fortunately, the payoff has been high (Kelley, 1999; Jones 2005). The costs
include a twelve-person spatial team, annual data costs, high spatial devel-
opment costs, hardware, and software. For instance, its investment of a few
million dollars for SST, its latest automobile navigation system (lower right
in Figure 7.2), resulted in GPS-referenced laptops being installed on 10,000
delivery trucks, which in turn improved delivery precision and efficiencies.
The benefits are increased productivity multiplied across a large workforce,
faster delivery response and optimized pathways, improved technician and
delivery person productivity, reduced overtime, lower support costs, and ca-
pability to optimally re-arrange service territories. Sears uses traditional fi-
nancial analysis with NPV, including computing payback periods. For very
large million-dollar-plus spatial projects, more advanced financial modeling
is done. Part of the challenge is to take into account integration of spatial
systems with legacy mainframe systems. This legacy integration is common
for GIS in large, data-rich companies. The company has measured huge pro-
ductivity and performance gains. The firm-value gains are perceived to be
there, but metrics do not yet break out the spatial contribution.
Prior to implementing GIS, the manual process at Sears was very time
consuming and wasteful, with many hours each day spent by routing-center
workers locating street addresses. In the early 1990s, an enterprise-wide GIS
system was constructed by geocoding Sears’ millions of customer addresses
and setting up optimized delivery with the goal of 90 percent reliability in the
promised delivery window (Kelley, 1999). The GIS calculates daily delivery
routes, based on a model that includes “estimated travel times, in-home time,
truck capacity, optimal stop sequence” (Kelley, 1999; Jones, 2005).
For tangible benefits, the system initially increased its efficiency by re-
ducing the time for routing and addressing from an average of 5 hours daily
to 20 minutes. The miles per delivery-truck stop were lowered by 0.6 mile,
which allows four more stops per truck per day (Kelley, 1999). Subsequently,
with further reduction in routing time, the system lowered the stops per truck
per day by another one-half (Jones, 2005). All this has allowed reduction in
the number of Sears national routing centers from forty-six to twelve (Jones,
2005). Delivery orders have expanded by 9 percent with the same-sized truck
fleet. In all, equipment and facilities savings from the GIS-based networking
enhancement are over $50 million per year (Kelley, 1999; Jones, 2005). All
this was accomplished with only small additions to IT staff to support GIS
(Jones, 2005).
The intangible benefits apply to Sears management and customers. Sears
middle and top management are able to use the information in this system
strategically to plan improved efficiency versus its competitors over long
periods of time, a strategic efficiency approach resembling Wal-Mart’s well-
known inventory and just-in-time delivery systems. Service management im-
proves, less customers are lost due to on-time deliveries; and competitive
advantage is gained with better service quality. At the customer level, re-
ducing the 20 percent of missed deliveries down to less than 5 percent has
enhanced the image for Sears Delivery which has also helped in the marketing
other company products.
The Sears Computer Aided Routing System (CARS) provides optimal rout-
ing for over 10,000 maintenance technicians servicing home appliances na-
tionally. The system decides on which specialists are routed to which sequence
of homes. Network algorithms group destinations into routes to minimize
travel time. As seen in Figure 7.2, two appliance generalists and one laundry
specialist each has a morning and afternoon route that has limited crisscross-
ing. The System is dynamic and can re-route if traffic blockages occur. This
happened with Appliance Generalist A, who at the end of the simulated day
was blocked by the upper bridge outage from getting from stop 6 to stop 7.
Instead he was dynamically re-routed across the lower bridge, taking a much
longer route, but getting to stops 7 and 8. It was a long day, but customers were
served!
In summary, for this enterprise-wide spatial system deployed in a large
corporation, the initial data acquisition was relatively expensive and time
consuming, since over four million addresses had to be geocoded and then
tediously corrected (Kelley, 2002). Another difference is the integration of
GIS with GPS on the delivery trucks. Knowing the location of each truck in
real-time allows more optimal management of the whole fleet. However, as
we discussed, the necessity to couple GIS with other systems may increase
the overall cost.
c07
Historical Demand
Current Demand
Forecasted Demand
Geography
Zip codes, travel time
Available Fleet
Vehicles, capacity,drivers
Work load balancing
GIS Based Service Area Planning

Sears Enhanced Home Delivery Warehouse Optimization using GIS
System (EHDS) Routing of forklifts within Sears
Nationwide GIS Street Level Distribution Centers to pick
Geocoding and Route Optimization merchandise for truck loading
October 18, 2007
Customer Request
Customer Service
Home Delivery of Merchandise
Delivery Vehicle departs to your home
OR to Deliver Merchandise
Home Product Repair
14:25
Service Technician Departs to your

Mainframe CIS home for Product Repair
217
GPS
In-Vehicle Navigation
Turn Right on East

Washington Street
Char Count=
Nationwide
Route Management Geocoding and Routing
Mobile Workstation
ArcObjects
Sears Computer Aided Routing System
Sears Mobile Mapping
Capacity Area Management System (CAMS) (CARS)
In-Vehicle Navigation
Planned Capacity of available Nationwide GIS Street Level
service technicians assigned to Geocoding and Route Optimization
geographic work areas
Figure 7.3 Sears Delivery—Transit Time Reduction. Source: Jones, 2005

THE PRODUCTIVITY PARADOX FOR GIS
In the information systems field, a special evaluation methodology has

evolved, referred to as the “value of IT investment” approach. The method
examines the level of IT investment and variety of returns on investment.
It is sometimes referred to as “productivity paradox,” i.e. the paradox that
companies or larger economic units invest in IT but fail to realize appropriate
productivity gains (Brynjolfsson, 1993).
This method is related to standard techniques of cost-benefit analysis.
The difference is that it is especially sensitive to IT measurement problems
in cost-benefit analysis. Hence, instead of the broad category of costs, it
focuses on investment in IT, which implies monetary funds intentionally
directed towards IT. Instead of encompassing the broad range of tangible and
intangible benefits, it focuses on quantitative measures of value, productivity,
and performance. Return on investment is regarded as best measured by multi-
attributes, not by a single attribute (Ahituv, 1980, 1989). This approach has
not solved the difficult problems of measuring benefits, since the difficulties
remain in deciding which attributes to measure, how to measure them, and
how to combine them in a multi-attribute function (Ahituv, 1989).
The relevance for GIS is that perceived value method is the most applicable,
since GISs are increasingly utilized for managing and decision-making, rather
than for low-level operations. Perceived value of information is valuation that
is entirely a subjective rating of the user regarding information value. The two
other types of valuation are normative value of information, i.e. a theoretical
model producing a quantitative value, and realistic value of information, i.e.
estimating differences in value by measuring before and after an information
system has been put in place (Ahituv, 1989; Ragowsky et al., 2000). At
the same time, the perceived value method has many more sources of error
and should be utilized with caution. Permanent problems with perceived
value are the following: (a) Individual respondents differ considerably in
their subjective reactions. (b) It is hard to translate subjective rating scales
into tangible dollar amounts.
In spite of these problems, the perceived-value approach can be applied to
analyzing the value of past, present, and future GIS systems. Care should be
taken to achieve a large enough sample, and to have a panel of questions that
may be partly convertible to tangible values.
A final issue in this section for the value of IT investment approach is that
of appropriate lag times for the value or productivity to be realized. Consider
the case of investing some of your own funds to purchase a laptop GIS system
for your individual use. How many months would be needed before you are
at the peak of realizing the benefits of the investment? The same issue of lag
time would occur for a large-scale enterprise GIS system. It may take many
years after the firm’s investment for such GIS payoffs to be realized (Devaraj
and Kohli, 2002).
CHAPTER SUMMARY 219
Overall, the value of IT investments methods offer many potential oppor-

tunities for persons responsible for assessing the benefits of GIS systems.
Because of the expanding investments being made by organizations in GIS,
these techniques should be seriously considered, to complement or replace
standard techniques of cost-benefit methods.
CHAPTER SUMMARY
Chapter 7 has examined the methods and procedures of cost-benefit analysis

for information technology, and sought to identify the distinguishing aspects
for GIS, compared to non-spatial IS. One major difference is that GIS has
high data-collection costs. This problem improved somewhat over time, as
web-accessible digital libraries have become available, often at lower cost
and improved quality.
Another area of relatively higher GIS costs is training and tech-
nical support, since training for new technologies may require extra
expenditures—internally, by web-based training, or through outside train-
ing services. University training offerings, including in business schools, are
currently limited.
The benefits for GIS may accrue somewhat later and be stretched out over
a longer period than for the average IS. This is because (1) organizations
often do not understand all the benefits where they are, and (2) benefits
often are enterprise-wide, and less visible by themselves. One benefit that is
frequently available is to sell GIS project results as a product. This needs to be
factored into the cost-benefit analysis at the beginning, even if it is somewhat
speculative.
This chapter leads to the following practical suggestions to GIS managers
and users on how to improve the quality of a GIS cost-benefit analysis:
r Follow a careful plan utilizing all the conventional C-B methods and
knowledge available for IS systems, as well as the more limited toolkit
for GIS.
r Gather as much post-audit benchmarking information as possible from
your organization.
r Give extra attention to estimating data acquisition costs. Examine what
are the sources for spatial data and, if applicable, how they can be inex-
pensively converted to digital. How will the database be organized with
its spatial and non-spatial components.
r Use a long enough cost-benefit timeframe that all the benefits can be
realized.
r Consider the cost of training. Anticipate the delays in realizing benefits.
r Do a careful analysis of intangibles and include them in the report to
management.
CASE STUDY: NORWICH UNION UNEXPECTED BENEFITS OF GIS
Norwich Union, one of the largest insurance firms in the UK, offers primary
lines of life, property, automobile, and casualty, but also mortgage loans, and
investment and health care products. Its GIS and GPS applications include
modeling of floods and other terrain, weather mapping, distributions of pen-
sion risks, online geocoding, automatic flood risk assessment for homes in
high flood-risk areas, and real-time spatial analysis of policyholders of auto
insurance who have GPS-enabled cars. A four-person GIS team develops and
implements spatial applications.
One of the two lead spatial products is Norwich’s flood map for the UK,
which is the country’s leading source of flood insurance information, both for
Norwich and other insurance companies that acquire the maps. The system has
recently been upgraded so an individual home can be assessed very precisely
online regarding the latest flood boundaries and an accurate rate can be given.
Due to this individual-home precision, the extent of flood zones has been
reduced.
The second lead spatial product, “Pay as You Drive” Insurance is a new one
that Norwich launched for young drivers in 2005. A GPS device is installed
in the car that estimates the customer’s monthly insurance premium, based on
when, where, and how often he/she drives. The data are transmitted from the
car of the policyholder to Norwich, via satellite (Figure 7.4). Based on the real-
time data and utilizing IBM software, Norwich adjusts the premium rates. This
is different from traditional car insurance, which calculates premiums yearly
based on actuarial tables. It’s analogous to mobile phone use, where a person
pays depending on usage. The young drivers have saved up to 25 percent on
their premiums, versus traditional. The product was rolled out to customers
in 2006. The benefit to Norwich is that the drivers with the devices tend to be
more cautious, and thus lower Norwich’s risk with this group of insured.
An unexpected plus is that the driving data for the driver population, with
the identities of policyholders stripped off, are useful to Norwich Union
internally for demographic and marketing studies. With appropriate legal
sanction, the stripped data could also become a product marketed to other
companies interested in driving and traffic patterns.
Norwich has a high level of net benefits, along with low costs, except
for purchase of expensive external data. What is different from the other case
studies is the collection of real-time spatial data on customer driving behavior,
and proportionate reductions in auto premium costs. This is made up for by
improvements in customer driving profiles through the ability of the customer
to assess his/her driving and secondly by Norwich Union’s potential to utilize
population data with identities stripped off, in-house and possibly for sale on
the outside. Along with the innovation associated with spatial technologies
comes the possibility to realize new and sometimes unexpected cost savings
and new areas of business. These savings and gains may be little known when
the cost-benefits are first calculated, but can be added to subsequent C-B
ACKNOWLEDGMENTS 221
Figure 7.4 GPS Device for Norwich’s “Pay as You Drive” Insurance. On top is
the GPS unit being plugged into the vehicle. The data are automatically transmitted
by satellite connection to Norwich Union. On the bottom, the driver can also plug
the device into his/her pc and review his/her driving behavior. Source: Stevenson and
Hunt, 2006
analyses in the later stages of product implementation once costs and benefits
can be better determined.
ACKNOWLEDGMENTS
Parts of this chapter were modified with permission from a book chapter by
James Pick in Geographical Systems in Business (James Pick, ed.), 2005,
Hershey, PA, Idea Group Publishing.
REFERENCES
Ahituv, N. 1980. “A systematic approach toward assessing the value of an information

system.” MIS Quarterly, December, 61–75.
Ahituv, N. 1989. “Assessing the value of information: problems and ap-
proaches.” Proceedings of the International Conference on Information Systems,
315–325.
Ahituv, N., and S. Neumann. 1990. Principles of information systems for manage-
ment. Dubuque, IA: W.C. Brown.
Ahituv, N., S. Lipovetsky, and A. Tishler. 1999. “The relationship between firm’s
information systems policy and business performance: a multivariate analysis.” In
Mahmood, M.A., and E.J. Szewczak (Eds.), Measuring information technology in-
vestment payoff: contemporary approaches, Hershey, PA: Idea Group Publishing,
62–82.
Allaway, A., L. Murphy, and D. Berkowitz. 2004. “The geographical edge: spa-
tial analysis of retail program adoption.” In Pick, James B. (Ed.), Geographic
information systems in business, Hershey, PA: Idea Group Publishing.
Aronoff, S. 1989. Geographic information systems: a management perspective.
Ottawa: WDL Publications.
Baily, M., and A. Chakrabarti. 1988. “Electronics and white-collar productivity.” In
Baily, M., and A. Chakrabarti, Innovation and the productivity crisis, Washington:
Brookings Institution.
Barua, A., C. Kriebel, and T. Mukhopadhyay. 1991. “Information technology and
business value: an analytic and empirical investigation.” Working Paper, Austin,
TX: University of Texas, May.
Bertschek, I. (Ed.): 2003. “Information technology and productivity gains and cost
savings in companies.” In Bertschek, I., New Economy Handbook, Amsterdam:
Elsevier, 213–249.
Boardman, A.E., D.H. Greenberg, A.R. Vining, and D.L. Weimer. 2001. Cost-benefit
analysis: concepts and practice, 2nd Edition. Upper Saddle River, NJ: Prentice
Hall.
Boyles, D. 2002. GIS means business. Volume 2. Redlands, CA: ESRI Press.
Breshahan, T.F., and S. Greenstein. 2001. “The economic contribution of informa-
tion technology: towards comparative and user studies.” Journal of Evolutionary
Economics, 11: 95–118.
Brynjolfsson, E. 1993. “The productivity paradox of information technology.” Com-
munications of the ACM, 36(12): 67–77.
Brynjolfsson, E., and L. Hitt. 1993. “Is information systems spending productive?”
In Degress, J., R. Bostrom, and D. Robey (Eds.), Proceedings of the Interna-
tional Conference on Information Systems, New York: Association for Computing
Machinery, 47–64.
Chau, P.Y.K. 1995. “Factors used in the selection of packaged software in small
businesses: views of owners and managers.” Information and Management, 29:
71–78.
Clarke, K. 2003. Getting started with geographic information systems. Upper Saddle
REFERENCES 223
Cron, W.L., and M.G. Sobol. 1983. “The Relationship between computerization and
performance: a strategy for maximizing the economic benefits of computeriza-
tion.” Journal of Information Management, 6: 171–181.
Davila de Icaza, A., and J.B. Pick. 1998. “Quantitative model of size and complexity
of prospective geographic information systems for regional governments in Mex-
ico.” Proceedings of Americas Conference on Information Systems, Atlanta, GA:
Association for Information Systems, 396–398.
Day, G.S., and P.J.H. Schoemaker. 2000. “A different game.” In Day, G.S., P.J.H.
Schoemaker, and R.E. Gunther, (Eds.), Wharton on managing emerging technolo-
gies, New York: John Wiley and Sons, 1–23.
Devaraj, S., and R. Kohli. 2002. The IT payoff: measuring the business value of
information technology investments. New York: Pearson Education.
Dickinson, H.J., and H.W. Calkins. 1988. “The economic evaluation of implementing
a GIS.” International Journal of Geographical Information Systems, 2: 307–
237.
Dickinson, H.J., and H.W. Calkins. 1990. “Concerning the economic evaluation of
implementing a GIS.” International Journal of Geographical Information Systems,
4: 211–212.
Directions Magazine. 2003. “Daratech reports GIS revenues forecast to grow 8%
to $1.75 billion in 2003; utilities and government increase spending.” Directions
Magazine, August 9. Available at http://www.directionsmagazine.com.
Doering, D.S., and R. Parayre. 2000. “Identification and assessment of emerging
technologies.” In Day, G.S., P.J.H. Schoemaker, and R.E. Gunther (Eds.), Whar-
ton on managing emerging technologies, New York: John Wiley and Sons, 75–
98.
ESRI Inc. 2003. Handheld product information available at www.esri.com.
Exler, R.D. 1988. “Integrated solutions for GIS/LIS data management.” GIS/LIS
[’]88 Proceedings, Volume 2, ACSM, ASPRS, AAG, URISA, 814–824.
Gerlach, J., B. Neumann, E. Moldauer, M. Argo, and D. Firsby. 2002. “Determining
the cost of it services.” Communications of the ACM, 45(9): 61–67.
Greene, R.P., and J.C. Stager. 2004. “Techniques and methods of GIS for business.”
In Pick, J.B. (Ed.), GIS in business, Hershey, PA: Idea Group.
Grimshaw, D.J. 2000. Bringing geographical information systems into business, 2nd
Edition. New York: John Wiley and Sons.
Harder, C. 1997. ArcView GIS means business. Redlands, CA: ESRI Press.
Harris, S.E., and J.L. Katz. 1989. “Predicting organizational performance using in-
formation technology managerial control ratios.” In Proceedings of the Twenty-
Second Hawaiian International Conference on System Science, Honolulu, HI.
Huxhold, W.E. 1991. An introduction to urban geographic information systems. New
York: Oxford University Press.
Huxhold, W.E., and A.G. Levinsohn. 1995. Managing geographic information system
projects. New York: Oxford University Press.
Jarupathirun, S., and F. Zahedi. 2001. “A theoretical framework for GIS-based spatial
decision support systems: utilization and performance evaluation.” Proceedings
of the Seventh Americas Conference on Information Systems, Boston, August,
245–248.
Jarupathirun, S., and F. Zahedi. 2004. “GIS as spatial decision support systems.” In
Pick, J.B. (Ed.), Geographic information systems in business, Hershey, PA: Idea
Group Publishing.
Jones, Steve. 2005. “Profiting from a Mobile Workforce GIS.” Paper presented at
ESRI Business GeoSummit, Chicago, April 19.
Karikari, Isaac, and John Stillwell. (2005). “Applying cost/benefit analysis to evaluate
investment in GIS: the case of Ghana’s Lands Commission Secretariat, Accra.”
Transactions in GIS, 9(4): 489–505.
Kauffman, R.J., and P. Weill. 1989. “An evaluative framework for research on the
performance effects of information technology investment.” Proceedings of the
Tenth International Conference on Information Systems.
Kelley, T. 1999. “Put your business on the map.” Transport Technology Today, April,
pp. 20–23.
King, J.L., and E.L. Schrems. 1978. “Cost-benefit analysis in information systems
development and operation.” ACM Computing Surveys, 10(1): 19–34.
Kingma, B.R. 2001. The economics of information: a guide to economic and cost-
benefit analysis for information professionals, 2nd Edition. Englewood, CO: Li-
braries Unlimited Inc.
Loveman, G.W. 1988. “An assessment of the productivity impact on informa-
tion technologies.” MIT Management in the 1990s Working Paper #88-054,
July.
Lucas, H.C. 1999. Information technology and the productivity paradox: assessing
the value of investing in IT. New York: Oxford University Press.
Maguire, David, Ross Smith, and Victoria Kouyoumjian. (2007). “Organizing for
Success: Building an ROI-based GIS business case.” Powerpoint slides from
Workshop at ESRI Users Conference, San Diego, June 26.
McCune, J.C. 1998. “The productivity paradox.” American Management Association
International, March, pp. 38–40.
Meeks, W.L., and S. Dasgupta. 2003. “Geospatial information utility: an estimation
of the relevance of geospatial information to users.” Journal of Decision Support
Systems, in press.
Meeks, W.L., and S. Dasgupta. 2004. “The value of using GIS and geospatial data
to support organizational decision making.” In Pick, J.B. (Ed.), Geographic infor-
mation systems in business, Hershey, PA: Idea Group Publishing.
Mennecke, B.E., M.D. Crossland, and B.L. Killingsworth. 2000. “Is A Map More
Than A Picture? The Role of SDSS Technology, Subject Characteristics, and
Problem Complexity On Map Reading and Problem Solving.” MIS Quarterly, 24:
4: 601–629.
Mitchell, A. 1999. The ESRI guide to GIS analysis, Volume 1: Geographic patterns
and relationships. Redlands, CA: ESRI Press.
Mueller, B. 1997. “Measuring ROI: can it be done?” AS/400 Systems Management,
November, 8–10.
Murphy, L.D. 1995. “Geographic information systems: are they decision support
systems?” Proceedings of 28th Annual Hawaiian International Conference on
Systems Sciences, Hawaii, 131–140.
REFERENCES 225
Navarrete, C.J., and J.B. Pick. 2002. “Information technology expenditure and in-
dustry performance: the case of the Mexican banking industry.” Journal of Global
Information Technology Management, 5(2): 7–28.
Navarrete, C.J., and J.B. Pick. 2003. “Information technology spending association
with organizational productivity and performance: a study of the Mexican banking
industry, 1982–1992.” In Shin, N. (Ed.), Creating Business Value with Information
Technology: Challenges and Solutions, 89–124.
Navarrete, C.J., and J.B. Pick. 2003. “Information technology spending and the
value of the firm: the case of Mexican banks.” In Shin, N. (Ed.), Creating
Business Value with Information Technology: Challenges and Solutions, 146–
165.
Niederman, F. 1999. “Valuing the IT workforce as intellectual capital.” Proceedings
of the SIGCPR Conference, Association for Computing Machinery, 174–181.
Noyelle, T. (Ed.): 1990. Skills, wages, and productivity in the service sector. Boulder,
CO: Westview Press.
Obermeyer, N.J., and J.K. Pinto. 1994. Managing geographic information systems.
New York: The Guilford Press.
Obermeyer, N.J. 1999. “Measuring the benefits and costs of GIS.” Chapter 42 in Lon-
gley, P.A., M.F. Goodchild, D.J. Maguire, and D.W. Rhind (Eds.), Geographical
information systems, Volume 2: Management issues and applications, New York:
John Wiley and Sons, 601–610.
Osterman, P. 1986. “The impact of computers on the employment of clerks and
managers.” Industrial and Labor Relations Review, 39: 175–186.
Phoenix, M. 2003. Personal communications based on ESRI Inc. Marketing Depart-
ment.
Pick, J.B. 1996. “GIS-based economic and social cluster analysis applied to a giant
city.” Proceedings of Americas Conference on Information Systems. Atlanta, GA:
Association for Information Systems, 515–517.
Pick, J.B., N. Viswanathan, and W.J. Hettrick. 2000a. “A dual census geographical
information systems in the context of data warehousing.” Proceedings of Americas
Conference on Information Systems, Atlanta, GA: Association for Information
Systems, 265–278.
Pick, J.B., W.J. Hettrick, N. Viswanathan, and E. Ellsworth. 2000b. “Intra-censal
geographical information systems: application to binational border cities.” Pro-
ceedings of European Conference on Information Systems, Vienna: Austria: ECIS,
1175–1181.
Pick, James B. 2005. “Costs and Benefits of GIS in Business.” In Pick, James B.
(Ed.), Geographic Information Systems in Business, pp. 56–79.
Pick, James B. 2006. “A Case-Study Analysis of Costs and Benefits of Geographic
Information Systems: Relationships to Firm Size and Strategy.” Proceedings of the
Americas Conference on Information Systems, in press, Atlanta, GA: Association
for Information Systems.
Ragowsky, A., N. Ahituv, and S. Neumann. 2000. “The benefits of using information
systems.” Communications of the ACM, 43(11): 303–311.
Reeve, D.E., and J.R. Petch. 1999. GIS, organisations, and people: a socio-technical
approach. London: Taylor and Francis.
Roach, S.S. 1989. “America’s white-collar productivity dilemma.” Manufacturing

Engineering, 104.
Sassone, P.G. 1988. “Cost benefit analysis of information systems: a survey of
methodologies.” Proceedings of Conference on Supporting Group Work, Associ-
ation for Computing Machinery, 126–133.
Satzinger, J.W., R.B. Jackson, and S.D. Burd. 2002. Systems analysis and design in
a changing world, Second Edition, Course Technology.
Smith, D.A., and R.F. Tomlinson. 1992. “Assessing the costs and benefits of geo-
graphical information systems: methodological and implementation issues.” In-
ternational Journal of Geographical Information Systems, 6: 247–256.
Sonnen, D., and H. Morris. 2005. “ESRI: Extending GIS to Enterprise Applications.”
White Paper, February, Framingham, MA: International Data Corporation.
Strassmann, P.A. 1997. “Computers have yet to make companies more productive.”
Computerworld, Sept. 15.
Strassmann, P.A. 1999. Information productivity. New Canaan, CT: Information
Economics Press.
Strassmann, P.A. 1999. Information productivity: assessing the information manage-
ment costs of U.S. industrial corporations. New Canaan, CT: The Information
Economics Press.
Swink, J., and C. Speier. 2000. “Presenting Geographic Information: Effects of
Data Aggregation, Dispersion, and Users’ Spatial Orientation.” Decision Sciences,
30(1): 169–195.
Tomlinson, R. 2003. Thinking About GIS: geographic information system planning
for managers. Redlands, CA: ESRI Press.
U.S. Census Bureau (2007). U.S. Census Bureau Website. Washington, D.C.: U.S.
Census Bureau. Available at www.census.gov.
Willcocks, L.P., and S. Lester. 1997. “In search of information technology pro-
ductivity: assessment issues.” Journal of the Operational Research Society, 48:
1082–1094.
Worrall, Les. (1994). “Justifying investment in GIS: a local government perspective.”
International Journal of Geographical Information Systems, 8(6): 545–565.
CHAPTER 8
MANAGING SPATIAL DATA
INTRODUCTION
The chapter concerns the design and uses of spatial databases and data ware-
houses. In the chapters on supporting business decisions and enterprise ap-
plications, databases and data warehouses were introduced as key constructs.
This chapter digs in deeper to gain fundamental understanding of these con-
cepts.
In the section on spatial data models, the relational, object-oriented, and
object-relational models are explained and compared on their pluses and
minuses. Situations appropriate for each are discussed. The case study of
the centralized spatial data repository at Enmax, an electrical and gas util-
ity firm owned by the City of Calgary, illustrates a success story in devel-
oping a modern, enterprise-wide spatial database in an inter-organizational
context.
The section on data warehouses builds on the introductory coverage in
Chapter 4 by discussing warehouse design, spatial functionality, and new di-
rections including the SOLAP model. An auto insurance firm and the City of
Portland illustrate alternatives in implementing spatial data warehouses. The
chapter ends by considering spatial applications’ data quality. This problem
may seem dull, but is highlighted when things go wrong, such as in trans-
portation emergencies, natural disasters, and insurance underwriting errors.
Data managers need to continually scrutinize for multiple sources of error in
spatial data.

227
228 MANAGING SPATIAL DATA
INTRODUCTION TO MANAGING SPATIAL DATA
Managing spatial data presents most of the issues of data management of non-
spatial data (Hoffer et al., 2006), but GIS-specific issues as well. A summary
of the key data management issues for MIS (see Table 8.1) shows that data
must be gathered, organized, checked for quality, made available to the users,
maintained, and updated.
A plan must be established to organize the data logically into a relational
or object-oriented data model. The relational model is based on organizing
data into a series of tables (Kroenke, 2005; Hoffer et al., 2006). For each
table, the rows represent records, while the columns indicate attributes. For
instance, XYZ Inc. has a relational table for its suppliers. Each row represents
a supplier, while each column represents name, e-mail address, number of
products sold to XYZ, total revenue of products sold per year to XYZ, and
date the supplier relationship started. In its overall relational design, XYZ
Inc. has tables for suppliers, parts, customers, employees, competitors, and
sales. These tables are related to each other by relational operators, each of
which causes logical operations on one or more of the relational tables such
as to join and divide, and select and query data items (Hoffer et al., 2006).
In an object-oriented data model, the key unit is the object, which represents
a real world “thing” having attributes and behaviors. It is able to relate to, and
communicate with other objects by sending messages to them that activate
their behaviors.
TABLE 8.1 Key Data Management Issues for MIS

r Organizational strategy for data management
r Data sources and gathering data
r Logical model for data
◦ Relational data model
◦ Object-oriented data model
◦ Data warehousing data model
r Data cleaning and quality checks
r Physical model for data storage
◦ Data-base management system
◦ Data warehouse
r Architecture that data storage resides in:
◦ Standalone
◦ Client-server
◦ Internet/web-based
r Data access versus data security
r User and customer training in data access
r Maintenance and updating of data
r Management, controls, evaluation
INTRODUCTION TO MANAGING SPATIAL DATA 229
The objects can be organized into hierarchical classes, so that characteris-

tics can be inherited from higher objects to lower ones. For instance, if the
class is car, the highest object is “Car” which has attributes which apply for
all cars. At the next level, the objects “Chevies,” “Toyotas,” “Mercedes,” and
“Volvos” inherit all the attributes for “Car,” but may have additional specific
attributes, for instance the “Volvo” adds the attribute, “ski compartment.” The
passing of attributes down the hierarchy is referred to as “inheritance.” There
are multiple other aspects of the object-oriented data model that are beyond
the scope of this book (Hoffer et al., 2006). The object-oriented model is
more suitable for models applied to rapidly changing environments having
complex behaviors.
Data warehousing has a different data model. Chapter 4 introduced it,
including how data enter and are processed in a data warehouse through the
extract, transform, and load (ETL) process. Inside the data warehouse, data are
logically organized into a central fact table that contains the most important
attributes as well as keys to access other tables. The topic is expanded on in
a major chapter section that emphasizes how data warehouses support spatial
analysis and applications.
Another aspect of data management is how the data are incorporated into
a network. They can be standalone, i.e. isolated from other systems, or in a
client-server array in which databases reside on one or more servers that are
connected to, and accessible by multiple client systems. Another arrangement
is the enterprise web architecture discussed in Chapter 1 and 5, for which data
located on web servers are accessible over the internet.There is frequently a
middle layer, often called middleware, between the web server client software
and the data-base or data warehouse. The middleware transfers processes and
data between the top and bottom layers and provides logic to control the
flows. This is often referred to as multi-tier, or n-tier architecture.
Other standard data management issues involve access and security, main-
tenance, customer training, and management (Hoffer et al., 2006; Dhillon,
2007; Laudon and Laudon, 2007). There is the tradeoff that tighter technical
security implies reduced access (Dhillon, 2007). This tug-of-war can only be
resolved by assessing the risks involved in security lapses versus those from
reduction of information flow to users and customers. Data maintenance,
training, data staffing, and training are other common operational practices
for information management (Laudon and Laudon, 2007).
Spatial data management utilizes all of the key MIS issues from Table 8.1,
but adds other ones specific to GIS and spatial. As seen in Table 8.2, GIS data
management issues (in bold) supplement the MIS issues.
The spatial layers introduced in Chapter 2 are built from digital points,
lines, and polygons. Combined together, the layers can produce a complex
map. However, the spatial data in one layer needs to be synchronized with
spatial data in the other layers in scale, coordinate system, and projection.
Spatial data can be stored in a variety of data formats, but to produce a map
they need to be converted to a common format.
TABLE 8.2 Key Data Management Issues for GIS (Issues in common between MIS
and GIS are shown by lighter shading)
r Organizational strategy for data management
r Data sources and gathering data
◦ Data compatibility for spatial layers
◦ Spatial data conversion
r Logical model for data
◦ Relational data model
◦ Object-oriented data model
◦ Object-relational data model
◦ Data warehousing data model
Accessing spatial attributes in the data warehousing model
SOLAP data model
r Data cleaning and quality checks
r Physical model for data storage
◦ Data-base management system
Spatial data-base management systems such as Oracle Spatial
◦ Data warehouse
Location of spatial data relative to the data warehouse
r Architecture that data storage resides in:
◦ Standalone
◦ Client-server
◦ Internet/web-based
Standards and protocols for web-based access to spatial data
r Data access versus data security
r User and customer training in data access
◦ User training to understand mapped and spatial displays of data
r Maintenance and updating of data
r Management, controls, evaluation
The standard non-spatial data warehouse model can be spatially-enabled.

This occurs in different steps in the ETL, inside the data warehouse, or in
external models that access it.
Another distinguishing aspect of spatial data models involves commercial
products that combine relational and object models. Leaders in the market
include Oracle 10g and the Geodatabase of ESRI’s ArcGIS.
The expanding area of spatial data management on the internet adds
standards and protocols specific for web-based data. In the management
realm, the biggest new spatial issue is managing security for the associ-
ated technologies of GPS, RFID, and sensors, and the enlarged issue of
user training, since often business users are unfamiliar with web mapping
concepts.
GIS, SPATIAL DATA, AND RELATIONAL DATABASES 231
The growth in the spatial sector of the economy, covered in Chapter 1, as

well as the greater internet and web-service delivery of spatial applications,
has brought with it a growth in the amount and volume of spatial data avail-
able worldwide. It is not only a question of choosing the best data model
and technical platform, but the explosion in information means that manage-
ment of content is more challenging. This adds another spatial management
challenge—managing the choice of data sources and also deciding how big
to grow a spatial database i.e. how to include as much relevant and afford-
able information as possible without overloading it. The manager needs to
know what data sources are available, their practical importance, what their
metadata are, their level of quality, cost, and update cycles.
GIS, SPATIAL DATA, AND RELATIONAL DATABASES
The leading model for spatial databases is the relational model. As mentioned
earlier, relational operators act on one or more relational tables to select,
combine, enlarge, or reduce tables through a sequence of operations. The
Project operator reduces a table to selected columns The Select operator
selects particular records based on the value(s) of an attribute. The Join
operator can combine two tables together based on common attributes. These
operators are illustrated in a sequence of operations in Figure 8.1. An initial
Volume/ Av.Day in Supplier Volume/ Av.Day in Supplier

Price Cost month inventory no. Price month inventory no.
10 7 1,000 45 18 10 1,000 45 18
50 30 100 150 75 50 100 150 75
150 85 9,500 25 1 Project 150 9,500 25 1
45 20 550 16 8 (4 columns) 45 550 16 8
75 65 754 185 22 75 754 185 22
200 125 300 56 11 200 300 56 11
Select Av Days in
Inventory > 50
No. of Credit Supplier
Name Customes Rating no. Volume/ Av.Day in Supplier
Price month inventory no.
Lopez 222 85 75
Join by 50 100 150 75
S&J 75 66 18
Supplier 75 754 185 22
Smith 300 45 15 No. 200 300 56 11
Vanh 15 91 22
No. of Credit Supplier Volume/ Av.Day in Supplier
Name Customes Rating no. month inventory no.
Lopez 222 85 50 100 150 75

Vanh 15 92 75 754 185 22
Figure 8.1 Example of Relational Tables and Sequence of Operations by Three

Relational Operators on Supplier Data
table of supplier information is reduced from five attributes (columns) to

four attributes (columns) by the Project operator. Next, only those suppliers
whose products have average days in inventory greater than fifty days are
chosen by the Select operator. Finally, that table is combined with a different
supplier table containing three new attributes (name, number of customers,
credit rating, and supplier number). The Join operator connects these tables
based on supplier numbers that are common between the two tables, that
is, Suppliers 75 and 22. The final table resulting from the Join consists of
complete attribute information from the two prior tables, but for only two
suppliers, Lopez and Vanh.
Other widely used operators are Union, which combines together the record
of two tables that have the identical attributes into one table. Difference
produces a third table that contains records present in the first table, but not
in the second one. Intersection of two tables results in a third table containing
the records that are common to both tables (Kroenke, 2005). Over twenty
other relational operators are available.
Spatial Relational Data Model

Although simple in its treatment of the spatial data, the relational model is
the most prevalent one for spatial data. It relates attribute data and spatial
data through an ID number. As seen in Figure 8.2, an attribute table’s ID
values have a one-to-one correspondence with spatial features. For instance,
for polygons at the top, polygon 3 corresponds to the ID value 3 in the table.
The second table in the figure on the middle right links four ID-field values
to four point locations, and the third table links five ID values to five line
segments. Besides “ID,” the tables contain other attributes. For instance,
if the upper attribute table represents records for sales regions, Attribute
A might be annual sales in the region, Attribute B number of stores in the
region, and Attribute C area of the region. If the second table represents
distribution centers, the attributes might be deliveries/day, floor area, number
of employees, and number of conveyer belts.
The spatial relational data model usually has the constraint that each at-
tribute table is associated with only one geometric form (point, line, or poly-
gon). Through the overlay of multiple layers, the resultant map can com-
bine the three geometric features as an end result, so any geometric form
can be represented. Figure 8.3 repeats the example of the relational opera-
tors from Figure 8.1 but now shows the point maps. Each point represents
a supplier. This clarifies how the relational data model can narrow down
or expand the geometric representation, while also changing the associated
attributes.
Think of this model applied to ten spatial layers with associated tables.
A set of relational operators working on each table could alter the set of
attributes and the features displayed for each map layer, leading to a complex
Figure 8.2 Relationship of Spatial Data and Attribute Data
map transformation. This small illustration demonstrates that this model can
support sophisticated spatial transformations.
On the spatial side of the relational model, the coordinate systems, map
projections, and scales of all the layers need to be adjusted to be the same, so
they overlay exactly. The unit of measurement (feet, meters, kilometers) also
needs to be identical for all layers. To keep track of projections and the other
standards, the metadata for each layer is maintained and needs to be accurate
and available.
The advantages and disadvantages of the spatial relational data model are
given in Table 8.3 (Tomlinson, 2003). Pluses of this model are tables that are
simple to understand and logical linkages between the tables and layers. The
model nicely supports query sequences based on attributes. However, it is
not as suitable for spatial analysis. As the leading spatial data model, it has
extensive datasets available and a large pool of skilled GIS people who can
utilize it.
Figure 8.3 Example of Relational Tables and Sequence of Operations by Three

Relational Operators on Supplier Data, with Spatial Layers Shown for Start and End
On the minus side, the layers of points, lines, and polygons are a vector
representation of the real world, i.e. they don’t support the raster-like detail.
The model is not naturally suitable to spatial analysis involving more complex
objects. Access speed is slower than for some other data models. Also, long
and complex sequences of table operations are challenging to understand and
program.
TABLE 8.3 Relational Database Model—Advantages and Disadvantages
Advantages
r Simple tables
r Linkage with spatial model simple
r User interface in software is simple and easy to read
r Tables, layers, and relational operators can be modified easily
r Data can be maintained independently of the application
r Well suited to query and analysis of spatial features
r Macros, tools, and wizards are fairly easy to program
r Is the worldwide standard, so extensive GIS datasets available
r Large group of skilled personnel to support development and use
Disadvantages
r Model provides limited representation of the real world
r Not naturally suitable for spatial analysis
r Slow query access, since data structures must be operated on each time there is access
r Complex data relationships (using relational algebra) are challenging to model
(Modified from Tomlinson, 2003)
Spatial Object-Oriented Data Model

The object-oriented data model is a very different approach to attribute and
spatial data than for the relational one. As seen in the left on Figure 8.4, an
object has a name, data attributes belonging to it, and the methods used to
process the object (sometimes called behaviors). It can be thought of as more
“active” than a relational table, since the methods perform calculations and
analysis for an object. The right side of the figure illustrates the object for a
fast food outlet. This outlet has non-spatial attributes (revenue, cost, etc.) as
well as spatial features of city and state. The methods can perform several
numerical calculations and merge spatial features.
The objects work together by relationships and communications. Relation-
ships describe the way objects are associated with each other. Examples of the
relationships of objects are cardinality and class hierarchy, as seen in Figure
8.5. The cardinality relationship indicates the ratios of number of instances
of one object to another. For instance, in the figure, there are many customers
for one fast food outlet, which is designated as M:1 cardinality. Cardinalities,
which can be M:1, 1:1, 1:N, and M:N, were explained in Chapter 6 in the
context of entity-relationship diagrams (see Figure 6.3). Cardinality relation-
ships help to understand how objects are related to each other for access
and retrieval of information. They also are useful in designing the optimal
arrangement of the database.
Figure 8.4 Spatial Object and Example
Customer Fast Food Outlet

Feature – point
Feature – point
Monthly revenue
Feature – ZIP code
Monthly cost
Name
Items served/mo.
Phone
No of employees
E-mail M 1 City
Calculate distance to Calculate profit

nearest Fast Food calculate taxes
Outlet Calculate productivity
Calculate no. of Maerge neighboring
customer in ZIP code fast food outlets
McDonalds Outlet Taco Bell Outlet
Distance from highway Burritos sold/mo.

McMuffins sold monthly Burritos revenues/mo.
Calculate impact of
highway on sales Calculate burritos
Calculate McMuffins as percent revenue/mo.
percent of items
Figure 8.5 Spatial Objects Showing Multiplicity and Class Hierarchy

Communications are achieved by sending messages to other objects car-

rying the name of a method (behavior) that the object is capable of doing.
A sophisticated object-oriented program has a sequence of messages flowing
between objects.
Another type of object relationship illustrated in Figure 8.5 is class hier-
archy. Classes represent different extents of generalization of objects. At the
top of a class is a very general object such as, say “business.” At the levels
underneath are more specific objects, say “retail business” and “camera store
business” at the next lower level. Each of those objects can be expanded into
more specialized objects at the third level, and so on. An object at a lower
level inherits the attributes from all the higher levels, plus it adds attributes
and methods from its own level. In the figure, “McDonald’s Outlet” inherits
all attributes and methods from “Fast Food Outlet,” such as “feature—point,”
etc. In addition, it adds two new attributes and two methods that are specific
to McDonald’s. Class hierarchy is helpful in organizing a group of objects
and avoiding repeating information.
For spatial analysis, the object model organizes spatial attributes and meth-
ods hierarchically. This allows easy identification of what spatial attributes
are generic and what are more specialized. For instance, if there are 200 fast
food outlets, they all have the attributes and behaviors from the top level,
but at lower levels, specialized additional spatial attributes can be added in
an organized way. In this way, a large group of related objects can be given
spatial properties. For instance, using the “Distance from Highway” attribute,
a spatial operator can identify easily every MacDonald’s within 1/8 mile of a
major highway.
The spatial object-oriented data model is still being developed and may
ultimately overtake the relational. Full coverage of the object model is
beyond the scope of the book. Its importance is partly that some of its
features can be combined in those of relational in an object-relational
model, which is present for some commercial packages and is covered
later.
The pluses and minuses of the object-oriented data model appear in
Table 8.4. Advantages stress that the object can represent more complex
things from the real world, that it’s intuitive, and has behaviors. Also, it
has multiple ways to organize objects, including cardinalities, classes, ag-
gregation, and composition (Kroenke, 2005; Hoffer et al., 2006). Spatial
attributes and behavior can be added to characterize individual and classes of
objects.
Among the minuses are that spatial object-oriented (OO) models are diffi-
cult to design (Tomlinson, 2003). In the real world, the methods (behaviors)
and communication sequences may be less well known. Since most databases
are relational, there are conversion issues in importing and exporting data.
Further, it is less likely that the GIS workforce knows OO models or is able
to program in OO languages (Tomlinson, 2003).
TABLE 8.4 Object-Oriented Data Model—Advantages and Disadvantages
Advantages
r An object has attributes and behaviors that give it identity and make it accessible
r The real world can have complex representations through multiple objects
r OO has an intuitive sense, since the objects are simplified forms of real-world objects
r Multiplicities (same as cardinalities) can be expressed
r There can be a hierarchy of classes with greater or lesser generalization
r Objects can be associated through aggregation and composition
r Certain object-oriented (OO) simulation languages are based on objects
r The behavior rules for objects can be used to maintain data integrity
r On the average, less user-code needed in GIS software to conduct the same query or
other user operation or series of operations
Disadvantages
r The complex OO models of the real-world are more difficult to design and build
r For the objects to mimic real world, the real world attributes, behavior, and
communication need to be known. This not always the case, for instance in nature
r Some business applications are not suitable to OO approach, for instance financial
models
r Large, complex OO models are difficult to execute
r Challenges of import and export with relational data-bases
r OO computer languages used to program the OO data model. There is less knowledge
among GIS workforce of OO computer languages
(Modified from Tomlinson, 2003)
Spatial Object-Relational Data Model

In the object-relational model, object-oriented capabilities complement a
relational database. Relational tables remain as the place for data storage,
but the relational model can interact with some object-oriented functionality
on top. This model is appearing in the commercial marketplace in varied
products including Oracle Spatial 10g and ESRI’s Geodatabase model of
ArcGIS, which is mixed object-relational. There is a question of which data
model is more suited for a specific business situation.
There are business situations where one or another of these major data
models is appropriate, as seen in Table 8.5 (Tomlinson, 2003). For example, a
utility firm has to store and model a large transmission network with numerous
assets such as power lines, transformers, poles, and distribution stations. It
needs to classify the relationships and inheritance of characteristics of these
assets. This is easily done through an object-oriented or object-relational data
model. In companies with flexibility and sufficient budget to choose between
models, the developer needs to analyze the primary logical characteristics of
the real-world attribute and spatial data, before selecting a data model.
TABLE 8.5 Appropriate Data Model for Certain Data Modeling Situations in Business
Logical data model Recommended

Data modeling situation characteristics logical data model
Retail chain that has store Simple relationships between Relational

locations, highways, and sales features
regions to do impact analysis of
vehicle traffic on store sales
Enterprise-wide application that Connection between large-scale Relational
connects data on a large relational data-bases
company’s sales with
business-partner data-bases
Small firm that needs to conduct Spatial analysis of data on Relational
side analysis with demographic demographics and competitors
and competitor data that currently is stored as a
non-spatial relational model
Energy utility that needs to Complex relationships with Object-relational or
model energy tranmission inherticance of attributes and object-oriented
network, including power lines, behaviors
transformers, poles etc.
Major petroleum firm must Multiple simultaneous updating Object-relational or
simultaneously updata multiple object-oriented
parts of its large pipeline
data-base with hourly additions
and repairs
An insurance firm has many Links to varied types of Object-relational or
legacy relational data-bases, relational and OO data-bases object-oriented
legacy relational spatial
data-base, and some
object-oriented database
(Modified from Tomlison, 2003)
Oracle Spatial 10g

Oracle Spatial 10g is a Spatial Relational Database that is a version of the
Standard Oracle 10 Database product, which is among the leading databases
for medium and large businesses. Besides advantages already discussed for
relational databases, it also has accessing and language features that are keyed
to the web and internet. Oracle Spatial 10g uses the open programming stan-
dards of SQL, XML, HTTP, HTML, Java, .NET, GML, OpenLS, ISO211,
and SVG (Lopez, 2005). The Java language is object-oriented, so OO pro-
grammers and developers can use it for internet programming.
Within Oracle Spatial 10g, there are a number of Spatial Functions available
(see Table 8.6). Among them is the geodetic function supporting the use of the
TABLE 8.6 Oracle Spatial 10g: Capabilities and Functioning of a Spatial RDBMS
STANDARD TYPES OF SPATIAL FUNCTIONS

SQL spatial type
R-tree index (a type of tree structure that can use coordinates)
Spatial operators (perform spatial analysis, leverage spatial indexes)
Spatial functions (perform spatial analysis, no leveraging of spatial indexes)
Geodetic support (Lat/Long)
Linear referencing
Spatial aggregates
Long transactions
Indexing, query, load
Partitioning
HIGHER LEVEL SPATIAL FUNCTIONS
GeoRaster Type
Network Data Model
Topology Data Model
Geocoding Engine
Routing Engine
Spatial Data Mining
Text Annotation Type
GML 2.0/SVG Support
Transportable Tablespaces
Reprinted with permission of the Geospatial Information & Technology Association, www.gita.org.
latitude/longitude coordinate system while other functions handle indexing,

partitioning, and aggregation for spatial data. Relational spatial operators can
change and transform spatial data.
The Spatial functions can determine the relationship of two spatial layers
to each other. Figure 8.6 shows a pair of layers and several functions to
manipulate them. Original refers to the starting layer, while Difference refers
to the layer paired with “Original.” Union includes both layers. Intersect is
the overlapping portion of the two layers when they are put one on top of the
other. XOR (exclusive OR) refers to the parts of the two layers that do not
intersect. The spatial functions are direct and useful.
Oracle Spatial 10g’s higher-level functions include the GeoRaster Type
that accesses raster imagery, network data model for spatial features that are
arranged in networks, topology data model for arranging spatial features in a
topology, a geocoding engine that geocodes address, and spatial data mining
that allows searching of spatial data. GML 2.0 is included that supports
the new GIS standards of GML and SVG for display of vector data on the
web. Overall the spatial analysis capability is moderate-level. However, for
mainstream business processing, the moderate-level spatial capability is offset
Figure 8.6 Spatial Functions in XML for a Spatial RDBMS. Modified from Lopez,
2006
by mainstream, highly efficient database processing capability for very large

scale data storage and fast transactions.
In Oracle Spatial, SQL, a structured query language, creates, retrieves,
updates, and deletes data. It is flexible for use on the internet. An example of
SQL for a spatial retrieval (query) of information is shown in Figures 8.7 and
8.8 (modified from Lopez, 2005). The user is asking: “What FedEx offices are
within 3 miles of my office?” Her office is located at 278 Corporate Avenue.
The map portraying this query is seen in Figure 8.7. In the center is the user’s
office, with a 3-mile ring around it. The SQL code to answer this query is
shown in Figure 8.8. This code asks to find all addresses from a list of Federal
Express offices where the distance between 278 Corporate Avenue and a Fed-
eral Express office is within 3 miles. “MDSYS.SDO WITHIN DISTANCE,”
a spatial function in Oracle 10g, asks, given the office point and distance
(radius), if the Fed Ex points are within that distance from the office point.
The example illustrates how SQL spatial functions can be utilized to perform
data retrieval.
The City of New York illustrates the full mainframe power of Oracle Spatial
10g. New York City standardized on this software, with the justification was
that Oracle had for some time supported the non-spatial, heavy-duty database
processing for the city (Lopez, 2005). The decision to centralize its spatial
applications on Oracle Spatial 10g as the main repository leveraged on the
Figure 8.7 Locational Query for a Spatial DBMS. Modified from Lopez, 2005
city’s existing Oracle knowledge and skill base, as well as offered the capacity
to support a very large spatial processing demand. As seen in Figure 8.9, the
data stored in Oracle Spatial 10g’s central repository include the topography
and cadastral information (i.e. land records), the physical assets of the city,
and transportation. Major business applications, shown in the right in the
figure, complement their own business data with spatially-enabled data from
the repository. These are mainstay city applications that previously ran non-
spatially under Oracle and are now spatially-enabled.
Another group of city departments, that includes environmental manage-
ment, crime monitoring, and power/water services, which appear on the left
in the figure, had previously run their own specialized GIS systems. They
continue to run specialized GIS software packages, but now can access at-
tribute and spatial data from the centralized Oracle 10g database. They use
specialized GIS software for spatial analysis and processing, since their local
Figure 8.8 Example of a Location Query in a Spatial RDBMS. Modified from

Lopez, 2005
Figure 8.9 Oracle Spatial 10g Integrated Data Architecture for New York City.
Reprinted with permission of the Geospatial Information & Technology Association,
www.gita.org.
GIS needs exceed the spatial analysis functionality of Oracle Spatial 10g.
This example demonstrates the huge mainframe scale possible for Oracle
Spatial 10g, as well as the flexible range of spatial sophistication that can be
provided to diverse city users.
The advantages of Oracle Spatial 10g (Lopez, 2005) are the potential for
high-volume spatial applications in the enterprise environment, and potential
in large IT shops to leverage the Oracle knowledge already present (see Table
8.7). A minus is that the GIS or IT department of a smaller enterprise may
not have the knowledge or skills to support Oracle Spatial. Also, the spatial
features are only moderate and the GIS interface may be less-friendly than
for some other packages. Perhaps the greatest deterrent is the high cost of
Oracle databases.
ESRI’s Geodatabases
ESRI’s Geodatabase in the ArcGIS software family is a way to query, store,
and process spatial information. At the low level, the data are stored in a
relational database (Rigaux et al., 2002), the logical design of which is seen
TABLE 8.7 Oracle Spatial 10g—Advantages and Disadvantages
Advantages
r The spatial and non-spatial information are stored in the same data-base which leads to
improved efficiencies.
r Spatial data can be centrally stored and easily integrated with enterprise and other major
business systems.
r Same SQL accessing method can be used for non-spatial and spatial data
r Utilities available for non-spatial are also available for spatial data including load,
import, export, and backup
r Advanced security and processing features of non-spatial Oracle are available to Oracle
Spatial, including security, replication, high availability, and parallel processing.
r For the internet platform, Oracle Spatial 10g uses open internet/web programming
standards such as Java, SQL, SML, and .Net.
r Many IT departments in medium- to large-sized businesses have Oracle skills sets that
can be leveraged for the Spatial version
Disadvantages
r The user interface for spatial features is more command-driven and less menu-driven
than some other alternatives
r The merger of relational data model for the main data-base with object-oriented
programming alternatives for the internet may add complexity for some developers
r GIS functionality within the spatial database is moderate and may not be suitable for
complex spatial applications
r For smaller enterprises, the Oracle-database training and skill sets may not be available
or affordable
(Modified from Lopez, 2005)
in Figure 8.10. Underpinning the Geodatabase are three key files, Relation,
Index, and Geometry. The Relation file is a standard relational table. The
Index file contains pointers, corresponding to the records in the Relation file,
that point to the physical location in the Geometry file of the spatial record,
which is variable-length. Relational operators can be used to join tables or
query for values.
Spatial Objects are available for point, line, polygon, and multi-point
(Rigaux et al., 2002). In additional to standard relational operators, the Geo-
database has structural elements that are summarized in Table 8.8. They
include some object-oriented features such as the Relationship Class, that
manages object classes. Feature Dataset collects together feature classes that
share the same spatial properties. This is a strong way to alter all at once
the spatial properties for a large group of geographic objects. Features are
also available to manage network connectivity and terrain modeling, and to
constrain the topology.
Figure 8.10 File Design to Represent a Theme in ArcGIS. Modified from Rigaux
et al., 2002
TABLE 8.8 Structural Elements of ESRI’s Geodatabase
Dimension (lengths or distances on map)

Feature class (same geometry type) e.g. point, line, polygon. Like features are stored
together, e.g. all ZIP code boundaries.
Feature dataset. A collection of feature classes stored together which share the same spatial
reference.”
Geometric network. A network where geometric coincidence implies feature connectivity.
Network dataset. The collection of all the network elements. These are elements for a generic
network including edges, junctions, turns, and others
Raster catalog. Collection of all the raster datasets defined in a table.
Raster dataset. Raster format organization.
Relationship Class. Manages relationships between objects in one class and those in another
class.
Schematic dataset. This dataset graphically represents network connectivity and sets of
relationships.
Survey dataset. Integrated collection of specialized survey feature classes and the
corresponding survey measurements.
Table. Relational table
Terrain. TIN-based dataset that uses feature classes to model using z-values. TIN
(Triangulated Irregular Network) is an ESRI terrain modeling application.
Toolbox. Collection of processes for analysis, modeling, and data management.
Topology. Constraints “on how point, line, and polygon features share geometry within a
geodatabase.” Topology “defines and enforces data integrity rules, topological relationship
queries and navigation, and sophisticated editing rules.”
Copyright © 2007 ESRI. All rights reserved. Used by permission.

ESRI’s Geodatabase offers a more elaborate and sophisticated spatial data

model than the vendor’s traditional shape file. It includes being able to apply
spatial relationships and integrity rules on a layer-by-layer basis (Arctur and
Zeiler, 2003). For instance, for a layer of “sales regions,” integrity rules could
be set so that no two sales regions can overlap. Another advantage is that
a layout for finished output can be generalized across multiple applications.
Geodatabase’s metadata feature is more elaborate and conforms to leading
standards such as ISO. Thus many organizations have a common way to
understand spatial datasets (Arctur and Zeiler, 2003). Among disadvantages
of Geodatabase is that it is more complex to understand and requires more
training and skill to use.
In summary, the relational-object data model has been illustrated by two
leading commercial products that utilize differently. The chapter completes
its coverage of spatial databases with a case study of a city-owned utility
company that developed and implemented a spatial database.
CASE STUDY OF ENMAX
Enmax is a private corporation wholly owned by the City of Calgary in

Canada. It serves a territory around Calgary of 422 square miles and has
over 360,000 customers. It distributes natural gas and electricity and has
started an initiative in wind energy. Its huge physical asset base includes
3,720 miles of gas distribution, 65,000 utility poles, 70,000 steel flag poles,
42,000 transformers, and thirty-seven substations, one of which is pictured in
Figure 8.11.
Before converting to enterprise-wide GIS, Enmax had cumbersome spatial
and locational procedures. An example is One-Call locates. This service
helps customers who need to know where Enmax-owned facilities are located
underground. It is needed by builders and developers who must avoid cutting
utility lines in construction projects. Enmax’s records on the location of
underground facilities were stored on microfilm and in paper map books.
Figure 8.11 Enmax Corporation Substation. Source: Enmax, 2007

CASE STUDY OF ENMAX 247
Since storing and accessing this information was slow and cumbersome,
One-Call was supported by twenty-five full-time staff.
The “hit rate,” i.e. the frequency of a One-Call customer digging up the
wrong location with its attendant hazards was a worrisome nine hits per 1,000
Locates. This and other problems led Enmax’s management to proceed with
a total overhaul of the spatial processing for the company. An enterprise-
wide spatial GIS was implemented that not only serves Enmax, but through
agreements serves other city and partnering agencies.
The Enmax configuration (see Figure 8.12) has the Oracle Spatial enter-
prise database at the center. As with New York City, the central repository
serves several categories of users: (1) Enterprise applications. These formerly
non-spatial business systems are being spatially-enabled where useful. (2)
GIS-enabled applications. These departments previously had some form of
GIS. Now most of their spatial and attribute data comes from the central
repository, although local specialist GIS software is used. (3) Drafting de-
sign and maintenance applications. This unit operates with Autodesk GIS
for drafting and construction, but that software is configured to access data
in the central repository. (4) Other spatial applications. These are mostly
partnerships between Enmax and other city agencies, regulatory agencies,
and private utilities. For instance, JUMP (Joint Utility Mapping Project) is a
formal collaborative agreement that sets the policies for spatial data sharing
Figure 8.12 Enmax Database Configuration. Modified from Lawrence, 2005

among Calgary’s water, sanitation, telephone, cable, and gas companies in-
cluding Enmax (Lawrence, 2005). Maps and spatial analysis can be added
to regulatory reporting documents. These applications increase the benefits
of the investment in GIS, in the form of new internal capabilities that ex-
pand Enmax efficiencies and performance, and spatial deliverables for the
partnership agreements and joint projects. For instance, outage management
and load flow analysis were formerly two separate GIS applications but now
access their data from the central repository. The cost savings from each is
estimated at $250,000 in one-time data-entry costs and $75,000 per year in
daily maintenance (Lawrence, 2005). In fact, benefits were realized across
the range of spatial applications.
Returning to the One-Call application and looking at it subsequent to
implementation of the enterprise spatial system, it is now supported by a
specialized GIS application that accesses data in the centralized Oracle Spatial
repository. Since One-Call Locates is part of the City of Calgary consortium,
Enmax’s underground GIS maps are shared with the other city utilities such
as telephone and water that also have subsurface assets. Using Field View
software, workers employ their laptops in the field daily to view the land,
estimate the sub-surface situation and update the centralized repository daily.
The improvements are large: One-Call Locates’ staff lowered from twenty-
five to five, while the error hit rate reduced from nine to one-quarter per
thousand. Its operating cost yearly lowered from $1.8 million (Canadian) to
$1.2 million. The better reliability and service times have resulted in much
higher service demand (Lawrence, 2005). Because of the joint agreement,
benefits are also realized by all the participating partners.
In summary, the Enmax case demonstrates how a bold move to a full
enterprise spatial database transformed business efficiencies and performance
at a mid-sized utility. It shows the benefits of locating corporate data in a
centralized database, reducing redundancies, and encouraging cross flows of
information internally as well as among cooperating partner organizations.
GIS AND DATA WAREHOUSES
Definition
The data warehouse, introduced in Chapter 4, is an alternative form of data
storage that differs from the conventional relational database. It is oriented
towards a subject-oriented view of data, rather than application-oriented. It
receives data from one or multiple relational databases, stores large or massive
amounts of data, and emphasizes permanent storage of data received over
periods of time. Data warehouses can be spatially-enabled in several ways.
The data in the warehouse can have spatial attributes, supporting mapping.
Mapping functions are built into some data warehouse packages. OLAP
“slicing and dicing” and what-if functions are performed on the data in
GIS AND DATA WAREHOUSES 249
the warehouse, and may include spatial characteristics. Furthermore, the data
warehouse can be linked to GIS, data mining, and other software packages
for more spatial and numerical analysis. Data warehouses and GIS used
conjointly emphasize the advantages of each, namely the large size, time
variance, and easy arrangement of data in the warehouse, along with the
spatial visualization and analysis capabilities of GIS.
Historical Background
It was not until the 1980s that data warehouses began to appear for use
(Gray, 2006). By 1990, the concepts of a data warehouse had developed
enough that the first major data warehouse textbook appeared (Inmon, 1990).
The analytical methods were a collection of methods drawn from statistics,
neural networks, and other fields. The theory of the processing steps for
data warehousing, OnLine Analytical Processing (OLAP), was formulated in
1995 by Codd (1995). The growth in the markets for data warehousing was
driven by the expanding data storage and its analytical uses in organizations.
During the past fifteen years, database companies such as Oracle and Sybase
produced data warehousing products as well as computer vendors Microsoft
and IBM, and ERP vendor SAP (Gray, 2006).
Fundamentals of the Data Warehouse

The principles of the data warehouse are covered first and then related to
GIS. A data warehouse differs from a relational database in the following
key characteristics: subject-oriented, time-variant, non-volatile, integrated,
and oriented towards users who are decision-makers (Gray, 2006, Gray and
Watson, 1997). Subject-oriented means that the user accesses information in
the data warehouse through common business subjects, such as part, customer,
competitor, order, and factory. This contrasts with relational databases that
often show the user many detailed attributes to access, but ones not necessarily
of high user importance. The traditional operational database focuses on the
functional areas of the business, such as sales, finance, and manufacturing.
The data warehouse keeps its older data, which makes possible analy-
sis of change tendencies over time. In other words, the warehouse data are
nonvolatile, i.e. after being stored, they are fixed over time. This lends sta-
bility to the data in a data warehouse (Inmon, 1990). The data warehouse
concept also favors the formation of summarized data, which are useful in
decision-making and also become nonvolatile. Granularity distinguishes the
individual data items from the summaries, i.e. the most granular refers to
raw data, while less granular is summarized data (Gray, 2006). An example
of summarized data is a summary of account totals for March, 2007, for a
business department.
The data warehouse is time variant. For the data warehouse, data are
extracted from multiple operational databases, made consistent, transformed,
scrutinized for quality, and then written to the data warehouse, with no further
updating allowed of those data, which is termed non-volatile. At the next
desired time point, data are again written to the data warehouse and become
non-volatile. Thus the data warehouse accumulates a time series of data by
extracting them at multiple time points. Hence the data are “time variant,” i.e.
they are available over a time period.
Data are extracted for the data warehouse from multiple sources, some of
which are legacy ones (Jukic, 2006), combined together, and written to the
data warehouse. Thus the data warehouse transforms the diverse data sources
into an integrated data set for permanent, long-term storage. Data are checked
for accuracy; and errors corrected.
This transforming is referred to as integration (Gray, 2006). The resultant
data, which are of high quality, diverse in sources, and extending over long
periods, are particularly suitable to research analysts and decision makers,
rather than operational transaction-based users.
In a large organization, data are gathered and transformed from a wide
collection of operational databases. The whole process of input, extraction,
error-checking, integration, and storing these data is known as ETL (extrac-
tion, transformation, and load) and was seen in Figure 4.9.
The data in the warehouse are organized by multiple dimensions and put
into a structure of dimensional modeling (Gray, 2006; Jukic, 2006). Di-
mensional modeling consists of arrangements of data into fact tables and
dimension tables. The fact table contains important numerical measures to
the user, as well as the keys to the dimension tables, which in turn include nu-
merical and descriptive attributes (Gray, 2006; Jukic, 2006). Spatial attributes
can appear in the fact table if they are key numeric facts for users, but are
more commonly put in dimension tables, where they can provide numeric
and descriptive descriptions, including geographic ones.
Two well-known types of dimensional models are the star schema and
snowflake schema (Gray and Watson, 1998). An example of a star schema,
shown in Figure 8.13, gives information on fast food sales and locations.
The fact table contains the keys to dimension tables and numeric attributes
on total sales and total managers. The location dimension table gives, for
each store, five geographic locations, ranging from county down to census
block. They can be used for thematic mapping and spatial analysis. Exact
point locations of stores (X-Y coordinates) could also be included, if deemed
important enough. The other dimension tables provide information on store
sales, products, and periodic reports.
GIS and spatial features can be present at several steps in the data ware-
house, shown in Figure 8.14, that elaborates on Figure 4.9 by indicating the
spatial features. The operational databases may be spatially-enabled. In that
case, the geocoding of data took place prior to the ETL process. Location
can be added as an attribute in the ETL step. Within the data warehouse, the
fact tables or dimension tables may identify spatial units, such as ZIP code or
county. The spatially-enabled tables may have address or X-Y coordinates.
Figure 8.13 Data Warehouse Star Schema, with Location Included
Geographical attributes would be located in the fact versus dimension table(s)

if location rose to high importance for the user. For example, in a data ware-
house of marketing data for sales regions, the region-ID is in the fact table.
GIS functionality is often present in some of the analysis and modeling
software packages shown on the right of Figure 8.14.
The most powerful functionality would be to process data from the data
warehouse with a full-featured GIS software package such as ArcGIS or
GeoMedia, which can perform a wide variety of GIS functions from over-
lays or distance measurement up to advanced features such as geostatistics,
modeling, 3-D visualization, and multimedia. It can enrich the uses of data
warehouse information for utilities infrastructure; energy exploration, pro-
duction, and distribution; traffic accident analysis; large scale auto insurance
risk analysis; management of fleets of vehicles; and business intelligence
for decision-making (SQL Server Magazine, 2002; Reid, 2006). As seen in
Chapter 4, GIS and data warehouses often serve as parts of an organiza-
tion’s enterprise architecture. They can function in a collaborative, coupled
environment with the other enterprise applications. A challenge is to connect
separate enterprise software packages together through robust and efficient
plug-in and connector software.
OnLine Analytical Processing (OLAP) is a set of rules for accessing and
processing multidimensional data in the data warehouse. Originally formu-
lated by E.F. Codd (1995), OLAP rules are focused on simple business
decision-making that directly accesses the dimensions of data in the ware-
house rather than on complex models. Among the main types of analysis are:
Figure 8.14 The Data Warehouse and Its Data Flows, Spatial Functions, and GIS
Components. Modified from Gray, 2006
(1) slice and dice, i.e. to divide complex datasets into smaller dimensions; (2)
drill down, to seek more detail in a report; (3) what-if changes for single or
multiple dimensions; and (4) access to the static, time-slice stores in the data
warehouse (Gray, 2006). OLAP is good at answering “why” and “what if”
questions.
Specifically, OLAP (OnLine Analytical Processing) refers to the following
characteristics of the information in the data warehouse (Codd, 1995): (1)
viewable in multiple dimensions, (2) transparent to the user, (3) accessible, (4)
consistent in its reporting, (5) based on client/server architecture, (6) generic
in dimensionality, (7) handle dynamic sparse matrices, (8) concurrent support
for multi-users, (9) has cross-dimensional operations, and (10) intuitive data
manipulation, (11) flexible in reporting, and (12) aggregation is possible.
Spatial data can be integrated into the OLAP model, which is termed
the SOLAP model (Bimonte et al., 2005; Malinowski and Zimanyi, 2003;
Marchand et al., 2003). The aggregation features of OLAP are modified
for SOLAP to handle geographic attributes. One approach is to modify the
OLAP’s multidimensional data model “to support complex objects as mea-
sures, inter-dependent attributes for measures and aggregation functions, use
of ad-hoc aggregation functions and n-to-n relations between fact and dimen-
sion” (Bimonte et al., 2005).
SOLAP models are still under development, in particular to formulate
improved SOLAP-based operators for spatial analysis, and more elaborate
working prototypes (Bimonte et al., 2005). In the future, a standard accepted
SOLAP model would allow OLAP’s what-if efficiencies for quick and flexible
access to multidimensional data in data warehouse to include the complexity
of spatial objects such as points, lines, and polygons. For some applications,
such a model might eliminate the need for standard GIS software packages.
Business intelligence. Business intelligence (BI) software packages often
have spatial features, as seen in Chapter 3. BI consists of interactive models
that are designed to assist decision-makers (Gray, 2006). In the context of
data warehouses, BI can conduct modeling based on information from the
data warehouse for forecasting, simulations, optimizations, and economic
modeling. Spatial capabilities can be present that include location in the
modeling and produce results as maps.
Data mining seeks to reveal useful and often novel patterns and relation-
ships in the raw and summarized data in the warehouse in order to solve
business problems. The answers are not pre-determined but often discov-
ered through exploratory methods (Gray, 2006). The variety of data mining
methods include intelligent agents, expert systems, fuzzy logic, neural net-
works, exploratory data analysis, and data visualization (Gray, 2006; Tan et
al., 2005). The methods are able to intensively explore large amounts of data
for patterns and relationships, and to identify potential answers to complex
business problems. Some of the areas of application are risk analysis, quality
control, and fraud detection.
There are several ways GIS and spatial techniques can be incorporated
in data mining. Before the data mining occurs, the data warehouse can be
spatially partitioned, so the data mining is selectively applied to certain ge-
ographies. During the data mining process, algorithms can be modified to
incorporate spatial methods. For instance, correlations can be adjusted for spa-
tial auto-correlation, and cluster analysis can add spatial indices (Viswanathan
et al., 2005). After data mining, patterns and relationships identified in the
data can be mapped with GIS software.
Physical Structure. Underneath the data warehouse’s conceptual struc-
ture, data are physically stored either in multidimensional databases keyed to
OLAP or in standard relational databases, which have slower performance.
The biggest vendors for physical data warehouses are Oracle, IBM, and Mi-
crosoft. Some of their products have built-in spatial functionality, e.g. Oracle
Spatial 10g.
Large data warehouses may store many terabytes of information, cost
several million dollars, and take up to two or three years to implement (Gray,
2006). Their pluses include better decision-making capability, faster access,
retention of data for longer time periods, and enhanced data quality (Oracle,
2006). Data quality is scrutinized and improved as part of the ETL process.
Spatially-Enabled Commercial Data Warehouses. Several major database

and ERP vendors including Oracle, IBM, and SAP offer spatially-enabled
database or data warehouse products. They are full-scale relational databases
or data warehouses that have spatial functionality built into them, including
spatial layers, geocoding, coordinate systems and projections, and spatial
analysis techniques. Although the functionality is not as elaborate as the
leading GIS software, it has gained in capability, to a level that satisfies many
everyday business needs for maps and spatial processing.
This following discussion introduces essentials on commercial spatially-
enabled data warehouses focusing on the Oracle Spatial 10g product. Earlier
in the chapter, Oracle Spatial 10g was examined relative to large-scale spa-
tial databases. However, Oracle Spatial 10g also supports the Oracle Data
Warehouse. The data warehouse product has features that include a multi-
dimensional OLAP engine and built-in ETL features. This emerging trend
demonstrates how existing mainstream enterprise packages can be commer-
cially modified for mid-level spatial applications, without the necessity of
connecting to traditional GIS software.
Design and construction of applications can be done through Oracle Ware-
house Builder, a graphical design tool having: (1) a graphical “design envi-
ronment” to create the data warehouse based on metadata, and (2) a “run-time
environment,” to convert the design into the physical processes that run the
data warehouse (Rittman, 2006). For viewing, Oracle Spatial 10g’s low-level
“Map Viewer” provides simple maps of the dimensional attributes and sum-
mary data in the data warehouse (Rittman, 2006). For higher-level spatial
analysis, major GIS vendor software such as ESRI’s ArcGIS or Integraph’s
GeoMedia can be applied.
Key Applications of Spatial Data Warehouses

Data warehouses and GIS are applied to large-scale datasets for analysis
of complex spatial problems that can include time. Important temporal ap-
plications are for market segmentation, insurance analytics, complex urban
transport, city traffic patterns and trends, patterns of credit results for regions
or nations, international tourism consumer patterns, financial fraud, consumer
loans (Gray, 2006; SQL Magazine, 2002; Reid, 2006). In this section, two
examples are given of real-world temporal applications: (1) auto insurance
applications (Bimonte et al., 2005) and (2) traffic patterns for the city area of
Portland, Oregon, over a time span of almost two decades (SQL Magazine,
2002).
Example of an Auto Insurance Application. Spatial data warehouses can
be built for large-scale analysis of auto insurance. In this example, the data
warehouse resides in Oracle Spatial 10g. The business items in the data
warehouse have location attributes that include census blocks, locations of
policies, business sites, landmarks, elevation, and traffic characteristics. For
data warehouses in auto risk insurance, maps can be produced that take spatial
views from the usual ZIP-code geography down to hundreds of block groups,
small areas within the ZIPs (Reid, 2006). This allows underwriters to set
more refined policy pricing. The geoprocessing needs to be fast, many tens
of millions of location data processed per day (Reid, 2006).
Example of a Local Government Application: City of Portland. The City of
Portland illustrates use of a customized connector program to connect a data
warehouse to a GIS. The data consist of city and regional traffic accidents from
the Oregon Department of Transportation. The solution combined an SQL
Server data warehouse with a customized program written in ArcObjects API
(application programming interface) from ESRI Inc. There is a pre-defined
schema of non-spatial and spatial attributes for transport of data between the
data warehouse and the ArcObjects program.
The city’s spatial data warehouse for city and regional traffic accidents has
over fifteen years of data and fourteen dimensions, including time, streets, age
and gender of participants, cause, surface, and weather. Following cleaning of
data entering the data warehouse, location coordinates are added by the GIS
team for each traffic accident. At the staging server, ETL extracts data weekly
to two powerful clustered production servers. When updates are called for, the
data warehouse repeats the ETL process, and outputs a new time slice of the
data.
The volume of data is huge, so attention was given to mitigating perfor-
mance bottlenecks (SQL Server Magazine, 2002). The solution included opti-
mizing replication of a time-slice of data, and partitioning the data warehouse
cube to speed up the average access time. Users of the city’s system utilize
interactive maps of all accident locations during the past decade and half, to
supplement accident reports and give added insight for decisions (SQL Server
Magazine, 2002). The data are stored in an SQL Server data warehouse.
The customized program allows the GIS software to utilize part or all of
the data warehouse. The City of Portland assigned programmers from its
Corporate Geographic Information System (CGIS) Group to program the
ETL and access modules, based on ArcObjects API for ArcGIS from ESRI
(SQL Server Magazine, 2002). Because of the scope of the programs, CGIS
limited the types of questions users can ask to a pre-defined set. The accident
outputs consist of tables and maps that are viewable on the web. Being able
to both query tables and visualize maps is regarded as crucial for users.
The benefits of this data-warehouse/GIS approach included halving of
replication time for a time slice of data, fast spatial queries, and response
times shortened by twenty-fold or more (SQL Server Magazine, 2002).
Future Directions in Data Warehouses

GIS and spatial technologies to data warehouse applications by providing
mapping and spatial analysis. Data warehouse applications can recognize
locations of organizational entities such as customers, facilities, assets, facil-
ity sites, and transport vehicles. GIS provides visualization and exploration
benefits to understand patterns and relationships of enterprise information
in the data warehouse, and support better decisions. The challenges are to
design spatially-enabled data warehouse architectures that provide added

value to corporate users and customers, are flexible enough to change with
the rapid technology advances in this field, and are efficient enough to achieve
satisfactory throughput and response times.
Future advances are anticipated that make data-warehouses more efficient,
improve integration of data warehouses with GIS, tune the analytic outputs to
the typical data-warehouse users, and coordinate the spatial data warehouses
with other enterprise software such as ERP, supply chain, and CRM. Since
the large datasets for each time-slice are written permanently to the data
warehouse, the location coordinates are often added in the ETL stage or
earlier. Future software needs to have more efficient tools available during
ETL such as geocoding, to minimize employee time spent.
There needs to be faster and more seamless connector software and in-
terfaces between commercial data-warehouse and GIS software. Analytic
software such as SAS and ArcGIS need to have analysis capabilities that
coordinate better with the data warehouse. The future shows promise that
SOLAP will become standardized as an extension of OLAP. Spatial data
warehouses serve analytic users who are interested in patterns, associations,
and longitudinal trends. In the future, the GIS and spatial functions will
hopefully be better focused to serve these users.
MANAGING DATA QUALITY
A general challenge for spatial data management is the quality of the data.
No matter how sophisticated the storage and access of data, for its ultimate
use, the data are only as good as their quality. An example from the field of
medicine is preventable deaths from medical errors, which was estimated as
44,000 to 98,000 Americans yearly (Institute of Medicine, cited in Pierce,
2003). Likewise with GIS, the impacts of poor data quality can sometimes be
profound. What if a governmental spatial system tracking shipments of nu-
clear materials has errors so that it recommends the wrong nuclear shipment
routes, compromising security. In business, what if an insurance underwriter
receives erroneous data from a spatial database about a large customer com-
mercial property and prices the property policy too low? What if a private
health-care firm’s ambulance routing software is inaccurate for section of
a city, cutting crucial minutes from the transport of critically ill patients?
Data quality is a crucial topic for the success of GIS. Management has the
responsibility to exercise control and maintain data quality.
Spatial applications depend on attribute and spatial data. Some generic
problems (Pierce, 2003) include: (1) multiple data sources i.e. there are in-
accuracies if a dataset is not collected consistently, (2) subjective bias for
qualitative data such as from questionnaires, (3) inconsistent data definitions
and formats, (4) slow data access through too high data volume or too slow
processors, (4) data accuracy problems, and (5) the conflict of security versus
MANAGING DATA QUALITY 257
accessibility. As regards the latter, if security is made too high, then the data
cannot be accessed which lowers the data quality. On the other hand, lax
security lowers the quality since data can more easily be compromised or
destroyed.
Data quality depends on the context of its use (Wang and Strong, 1996;
Pierce, 2003). The nature of the problem being solved determines what is the
appropriate data accuracy, speed of access, and completeness. For example,
for the digital layer of street boundaries in the Phoenix metropolitan area,
data quality criteria vary between a construction company that is re-building
street curbs versus a fast-food company estimating numbers of car passengers
passing by their outlets. The construction firm requires high ground accuracy
with resolution in inches, while the fast-food company needs lower accuracy
and resolution to assess traffic flows. A real estate company would see it
differently. Another example is completeness of housing data. A marketing
firm only needs a sample of the data, while city planning needs information
on every house. Hence, there is no firm and fast data quality standard. The
key questions are, who is the user? How does the user utilize spatial outputs?
In specifying it, an organization needs to explain context as well.
Spatial data have some distinctive aspects of data quality (Kearney, 2003;
Tomlinson, 2003). They include:
1. Spatial completeness. Are there sufficient types and numbers of spatial

features for the problem at hand?
2. Coverage. Does the geographic extent of the data correspond to the
extent of the problem at hand? Are the geographic features consistent
in the procedures used to locate them across the whole coverage? For
instance, a coverage having some points entered recently by GPS and
others by scanning an old map is not consistent. The coverage concern
is harder to address if the data provider didn’t provide the metadata.
3. Transforming spatial data. When data are aggregated, joined, split apart,
and queried in the data transformation inside databases and data ware-
houses, errors can occur leading to erroneously transformed results.
4. Accuracy. This can be divided (Tomlinson, 2003) into referential (error
in referring to a spatial feature), topological (error in the presenting of
the topology, such as a broken line segment), relative (two features are
not located correctly one to the other), and absolute (error in the map
position relative to the true earth position).
For real-time spatial applications such as GPS-based trucking systems that

are continually tracked, data quality is impacted by fast response time. Users
expect to have the data within seconds. If it takes minutes to get answers
about where the fleet is located, efficiency can be reduced and data quality
lowered. Quality is also affected by data conversion. If data are transformed
between departments or organizations, data quality may be reduced. As an
example, a gas utility company converts geometric data from use in CAD
(computer-aided design) to use in GIS for modeling network connectivity
(Dangermond, 2006). For the firm, the conversion is expensive and lengthy,
and many errors occur in the process.
Data quality is best planned as part of systems analysis and design from
the start. It is a serious management oversight to not address it early but wait
until users start complaining about mistakes based on faulty data produced
by the system. Even if designed into a GIS, data quality is contextual so
operating managers need to examine how the environment for the GIS is
changing internally and externally, who are the current users, and what are its
data-quality impacts.
CHAPTER SUMMARY
Data management is essential to GIS success. The chapter covered the fun-
damentals of spatial databases including its major types of relational, object-
oriented, and object relational. Each data model type has pluses and minuses
and is appropriate for certain problems. Two leading commercial data mod-
els from Oracle Spatial 10g and the Geodatabase of ArcGIS are both more
complex than earlier models and more flexible for a variety of applications.
The Enmax case study demonstrates how a utility can be transformed
through radical adoption of an enterprise-wide central data repository. The
case had the additional aspect of showing benefits of inter-organizational
sharing of a spatial database.
Data warehouses contrast with databases in being non-volatile and storing
data historically. The data warehouse has a variety of possible places for
spatial enablement. The examples of the City of Portland and an auto insurance
firm demonstrate where the data warehouse may be preferred over databases.
The data quality issues permeate the chapter, since the use of data is
compromised if quality is low. Data quality must not only be an early goal
in GIS system development, but continue right on through for implemented
systems.
ACKNOWLEDGEMENT
The section on GIS and Data Wavehouses was published in an ntry in Ency-
clopedia of Geographical Information Science “Data Wavehouses and GIS.”
by same B pich. It is used with the permission of Springer Verlag.
REFERENCES
Bimonte, S., A. Tchounikine, and M. Miquel. 2005. “Towards a spatial multidimen-

sional model.” In Proceedings of the 8th ACM International Workshop on Data
Warehousing and OLAP, November 4–5, Bremen Germany, pp. 39–46.
REFERENCES 259
Codd, E.F. 1995. “Twelve Rules for On-Line Analytic Processing.” Computerworld,
April 13.
2006 Proceedings.
Wiley and Sons.
Gray, Paul. 2006. Manager’s guide to making decisions about information systems.
Hoffer, J.A., M. Prescott, and F. McFadden. 2006. Modern Database Management,
8th Edition. Upper Saddle River, NJ: Prentice Hall.
Inmon, W.H. 1990. Building the data warehouse. New York: John Wiley and Sons.
Intergraph. 2003. “Toward a Spatial Information Repository Act.” Madison, AL:
Intergraph Mapping and Geospatial Solutions.
Jukic, N. 2006. “Modeling strategies and alternatives for data warehousing projects.”
Communications of the ACM, 49(4) 83–88.
Kearney, Pat. 2003. “Looking for Quality Geographic Data?” GeoWorld, 16(9):
38–40.
Keenan, Peter B. 2006. “An Electronic Market for Spatial Data.” Proceedings
of the Americas Conference on Information Systems, Acapulco, Mexico, Au-
gust 4–6. Atlanta, GA: Association for Information Systems. Available at
http://home.aisnet.org.
Kramer, Gail, and Stuart Nisbet. 2006. “The Practical Union of OLAP Analysis and
Geographic Mapping.” ArcUser, 9(1): 34–37.
Kroenke, David. 2005. Database Processing: Fundamentals, Design, and Implemen-
tation. 10th edition. Upper Saddle River, NJ: Prentice Hall.
Lawrence, Brad. 2005. “Spatial Databases.” Presentation at Geospatial Information
and Technology Association Conference 28, Denver, CO, March 6. Aurora, CO:
Geospatial Information and Technology Association. Available at www. gita.org.
Lopez, Xavier R. 2005. “Spatial Databases.” Presentation at Geospatial Information
and Technology Association Conference 28, Denver, CO, March 6. Aurora, CO:
Geospatial Information and Technology Association. Available at www. gita.org.
Lopez, Xavier R. 2006. “Spatial Data Warehouse: Enabling Enterprise-wide Decision
Support.” Proceedings of the Geospatial Information and Technology Association,
Conference 29, Aurora, CO: Geospatial Information and Technology Association.
Available at www. gita.org.
Malinowski, E., and E. Zimanyi. 2003. “Representing spatiality in a conceptual
multidimensional model.” In Proceedings of the 12th annual ACM international
workshop on geographic information systems. Washington, DC: ACM Press,
pp. 12–22.
Marchand, P., A. Brisebois, Y. Bedard, and G. Edwards. 2003. “Implementation
and evaluation of a hypercube-based method for spatio-temporal exploration and
analysis.” Journal of the International Society of Photogrammetry and Remote
Sensing, 59:6–20.
Oracle. 2006. Oracle Data Warehousing. Available at www.oracle.com.
Pick, J.B., N. Viswanathan, and W.J. Hettrick. 2000. “A dual census geographi-
cal information system in the context of data warehousing.” Proceedings of the
Americas Conference on Information Systems. Association for Information Sys-
tems, Atlanta, GA, pp. 265–278.
Pierce, Elizabeth M. 2005. “Introduction.” In Wang, Richard Y., Elizabeth M. Pierce,
Stuart E. Madnick, and Craig W. Fisher (Eds.), Information Quality, Armonk, NY:
M.E. Sharpe, pp. 3–17.
Reid, Hal. 2006. “Applying Oracle Spatial to a very large insurance problem.” Loca-
tion Intelligence, June 20. Available at www.locationintelligence.net.
Rittman, M. 2006. “An introduction to Oracle warehouse builder 10g.” DBAzine.com,
August 19. Available at www.dbazine.com.
Rittman, M. 2006. “ GIS-enabling your Oracle data warehouse.” DBAzine.com, April
18. Available at www.dbazine.com.
Sonnen, David, and Henry D. Morris. 2005. “Oracle 10g: Spatial Capabilities for En-
terprise Solutions.” IDC White Paper, February, Framingham, MA: International
Data Corporation.
SQL Server Magazine. 2002. “City of Portland Tames Massive SQL Server Data
Warehouse.” SQL Server Magazine, June 27. Available at www.sqlmag.com.
Tan, P-N, M. Steinbach, and V. Kumar. 2005. Introduction to data mining. Upper
Saddle River, NJ: Addison-Wesley.
Tomlinson, R. 2003. Thinking About GIS: geographic information system planning
for managers. Redlands, CA: ESRI Press.
Viswanathan, Nanda. 2005. “ GIS in Marketing,” in Pick, James B. (Ed.), Geographic
Information Systems in Business, Hershey, PA, Idea Group Publishing, 236–259.
Wang, Richard Y., and D.M. Strong. 1996. “Beyond Accuracy: What Data Quality
Means to Data Consumers.” Journal of Management Information Systems, 12(4):
5–33.
Wang, Richard Y., Elizabeth M. Pierce, Stuart E. Madnick, and Craig W. Fisher
(Eds.). 2005. Information Quality. Armonk, NY: M.E. Sharpe.
Watson, Hugh, and Paul Gray. 1997. Decision Support in the Data Warehouse. Upper
CHAPTER 9
ORGANIZATIONAL AND INDUSTRY

ASPECTS
GIS is implemented in businesses in a variety of organizational settings. For

instance, a real estate firm is organized to have a very small GIS department,
provide web-based access, and give responsibility for spatial use to highly
paid sales agents nationwide. A large utility holding firm with many sub-
sidiary companies has a corporate-wide department of twenty skilled GIS
staff that plans strategically, implements enterprise-wide, and coordinates
with small decentralized GIS departments in the subsidiaries. Industries dif-
fer in their organizational emphasis on spatial. While GIS is part of corporate
culture in highly strategic industries such as transportation, typically it has a
service-support function in health-care firms. In utilities, spatial technologies
are transitioning from centralized and mainframe-based systems to field de-
ployment by staff who are becoming empowered by spatial information and
the ability to support spatial decisions in the field.
This chapter begins by examining GIS and the organization, drawing on
results from the book’s 20 research case studies and on published research
studies to a limited extent. A general model of maturity stages of GIS by
industry is presented that conforms to Nolan’s six stages of IS growth from
Chapter 3. From that classification, three of the most important industry sec-
tors for GIS are discussed in this chapter and five others in Chapter 10, namely
insurance, marketing/geodemographics, banking and finance, healthcare, and
agriculture. It next turns to GIS in three spatially mature and leading indus-
tries of utilities, transportation, and retail. Due to space limitations, Chapters

261
262 ORGANIZATIONAL AND INDUSTRY ASPECTS
9 and 10 cannot cover the whole array of industry sectors, but some examples
of the other sectors appear elsewhere in the book.
GIS AND THE ORGANIZATION
So far, relatively little research or business attention has been paid to the
organizational aspects of GIS, while the differences between industry sec-
tors are an important issue in business. Organizational structure is practically
significant because GIS is an integrated function that needs to have organi-
zational connections with IT, corporate planning, marketing, and customers.
No matter if a company is small or large, outcomes depend on whether GIS
staff and services are hierarchical or flat, centralized or decentralized, whether
there is able leadership and/or good interpersonal communications, and how
motivated are GIS developers, workers, users, and customers. To illustrate
organizational factors, the Large Personal/Corporate Bank (LCPB) discussed
in Chapter 3 has a decentralized approach to GIS. As part of a global bank
headquartered in England, the U.S. division has a small GIS team that focuses
on location and trade-area analysis for branch expansion. In the Mexican di-
vision, there is a much larger GIS group that is more technical and does
more development and programming. Since the national units of the bank are
distinctive, it makes sense that the GIS national goals are also distinctive.
People working in GIS are perhaps its most valuable resource, yet that side
of geo-business has less known about it. Since GIS in business only gathered
momentum within the last fifteen years, the workforce is heterogeneous and
often multi-disciplinary (Marble, 2006). It is so young as a field that it has
not yet reflected on or assessed the workforce and skills, although the U.S.
Department of Labor has funded a project to do just that (Marble, 2006;
Wachter, 2006). Since the spatial arena covers many different technologies,
skills, and sciences, it’s a challenge just to bound the set of skills involved,
much less possess working knowledge of them (Marble, 2006).
Marble (2006) provides the following draft definition of the geospatial
workforce:
“The geospatial industry engages, at a variety of spatial scales, in the acquisi-

tion, integration, analysis, visualization, management, and distribution of data
having an explicit spatial and temporal context. A critical component of the
industry involves the design, construction, and testing of both hardware and
software tools to support these activities. Because of the highly technical nature
of the industry, it is also essential to include in any definition those organiza-
tions and individuals in the higher education community who are engaged in
relevant instructional activities.”
This points to workers who have a mixture of IT, geography, management,

and spatial analysis background and skills. The variety is also reflected in a
GIS AND THE ORGANIZATION 263
TABLE 9.1 Twelve Job Roles of GIS Professionals
Applications Identify and develop tools and instruments to satisfy customer needs
Development
Data Acquisition Collect geospatial and related data
Coordination Interorganizational facilitation and communication
Data Analysis and Process data and extract information to create products, drive
Interpretation conclusions, and inform decision-making reports
Data Management Catalog, archive, retrieve, and distribute geospatial data
Management Efficiently and effectively apply the company’s mission using
financial, technical, and intellectual skills and resources to
optimize the end products
Marketing Identify customer requirements and needs, and effectively
communicate those needs and requirements to the organization, as
well as promote geospatial solutions
Project Effectively oversee activity requirements to produce the desired
Management outcomes on time and within budget
Systems Analysis Assess requirements to produce the desired outcomes on time and
within budget
Systems Integrate resources and develop additional resources to support
Management spatial and temporal user requirements
Training Analyze, design, and develop instructional and non-instructional
interventions to provide transfer to knowledge
Visualization Render data and information into visual geospatial representations
Source: Gaudet et al., 2001.
list of twelve job roles for geospatial professionals seen in Table 9.1 (Gaudet
et al., 2001). The roles encompass data collection, management, marketing,
systems, training, and visualization. Appropriate education pathways are di-
verse. Marble (2006) points to the critical roles that university education and
commercial training will play in further developing an enlarged geospatial
workforce. He emphasizes a need for universities to restructure and redesign
their spatial programs and curricula.
The twenty research case studies included in this book reflect the organi-
zational differences and variety of skills and background of the workforce
involved. Regarding the organizational structure, the surveys yielded the fol-
lowing summary results.
r For large- and middle-sized enterprises, GIS tends to be located in its

own corporate department separate from IT.
r Some middle-sized firms have GIS located in a marketing department.
r For large- and middle-sized enterprises, GIS does not have a staff repre-
sentation at the corporation’s executive level.
r Technical skills for spatial projects tend to be in the GIS group but are
associated with IT skills in the IT group, or in some cases IT personnel
loaned to GIS.
r Outsourcing of GIS workforce is not present in the twenty firms, although
two firms (Engineering Systems and URS) provide consulting and par-
tial outsourcing largely to government and somewhat to business. The
outsourcing topic is returned to at the end of the next chapter.
r For most of the firms that had both GIS and IT groups, the amount
of collaboration and teamwork between the two groups varies between
extremes of strong to weak, and occasionally is contentious.
r For large- and medium-sized firms, GIS workforce tends to interact the
most with people and customers in the key spatial application areas.
For instance, at Lamar Advertising, GIS people interacted the most with
the national sales department, which constitutes the most important user
group.
r For small firms, GIS is a major part of the entire business. The key firm
leaders were experienced and knowledgeable in it from management
and/or technical aspects. Most of the workforce of these firms is spatially
skilled. There are neither separate departments for GIS or IT nor is the
distinction between GIS and IT evident organizationally or behaviorally.
Based on the interviews of twenty firms in the research case study sample,
larger companies implementing GIS and spatial technologies tend to have
specialized GIS departments. The departments are staffed by people with the
wide variety of skills, education, and experience backgrounds that is typical
of the geospatial workforce (Marble, 2006).
The organizational structure of GIS has been little researched, although
one study of Indian regional governments implementing GIS pointed to the
importance of organizational networks (Walsham and Sahay, 1999). Given
the importance of having the right organizational structure and people, these
areas hopefully will be studied more. The knowledge gained will be useful
to business managers. It can also be included in GIS education and training.
One goal of this book is to provide such practical knowledge.
GIS IN INDUSTRY AREAS
This chapter section examines GIS and spatial technologies in three leading
industry sectors. Its objective is to gain general understanding of the spatial
applications for particular industries and examine industry-specific versions
of GIS applications. GIS plays significant roles for the lead industries, so
a short summary appears at the beginning of each industry section of the
factors that distinguish GIS for that industry. Some examples from these lead
GIS IN INDUSTRY AREAS 265
industries have been given in prior chapters and cases. The difference here is
that industry features are the focus.
As pointed out in Chapter 3 for SDSS, GIS in business tends to be catego-
rized in early stages of growth. SDSS is entering an expansion stage spurred
by reduction in processing costs and the popularity of the web enterprise
platform and consumer spatial web products. The level of spatial decision
support in businesses tends to be at Nolan’s Stage 1 (Initiation) or Stage 2
(Contagion).
In this chapter, spatial maturity stages are assessed by industry. In some
industries spatial technologies have reached Stage 3 in which rapid growth is
slowing down and management controls and accounting scrutiny are being ap-
plied. A few industries have progressed to Stage 4 by developing standard use
of the spatial technologies and integration of them with other business systems
and procedures. Based on the opinions of a group of four experts on sector
maturity from the GIS industry, a classification of stage-of-growth maturity
for twenty-five leading industries in the private sector is presented in Table
9.2. The experts gave subjective ratings of industry-sector classification based
on what they considered the average GIS maturity level of firms in the sector.
The composition of industries is modified from a list by Hoovers Inc.
(Hoovers, 2006). Four industries, Defense and Homeland Security, Energy
and Utilities, Transportation, and Natural Resources and Environmental are
classified in Stage 4 (Integration). They are some of the industry sectors that
first utilized GIS and are among the largest today in spatial market size. The
next group in Stage 3 (Control) includes Communications, Insurance, Mar-
keting/Advertising, Oil and Gas, and Retail. They have become somewhat
regularized with management controls in place, but are less integrated with
other business systems and functions. Ten industries are categorized in Stage
2 (Contagion) which implies users are demanding much more use of them
and they are in a growth phase. For instance, spatial applications are growing
rapidly in parts of the banking industry, including for siting, CRM, and branch
management, market analyses. Lastly, six industries, apparel, chemicals, elec-
tronics, food and beverages, manufacturing, and publishing, are categorized
in the Stage 1 (Initiation). These industries offer instances of GIS and spatial
uses, but they haven’t yet taken off with rapid adoption and diffusion (Rogers,
1995). The industries tend to be more capital-intensive, which can hold back
rapid technological advance due to capital budgets that are often tight, the
cyclical nature of their business, and in some cases traditions and/or unions
that resist new technologies.
Although the industries have been placed in these categories, individual
firms vary within an industry. The real estate firm of Sperry Van Ness from
Chapter 1 brought GIS into widespread and critical use, advancing it beyond
the real-estate industry norm of Contagion Stage into the next stage of Control.
Almost all industries in the classification are becoming more mature, so it is
anticipated the chart entries will move to the right over time.
c09
TABLE 9.2 Industries in the U.S. Classified Nolan’s Stage of Growth for GIS/Spatial Technologies
Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Stage 6

Initiation Contagion Control Integration Data Administration Maturity
Defense and Homeland Security

Energy and Utilities

Transportation
Natural Resources and Environmental
Communications
Insurance
Marketing/Advertising
Oil and Gas
October 19, 2007
Retail
Agriculture
Banking and Finance
12:50
Construction
266
Education
Health Care
Leisure
Metals and Mining
Char Count=
Newspapers
Pest Management
Real Estate
Apparel
Chemicals
Electronics
Food and Beverages
Manufacturing
Publishing
Note: Industry Classifications modified from Hoovers Inc., 2006.

Original Source for Stages of Growth: Nolan, 1979.
UTILITIES 267
The eight industries selected for intensive coverage in Chapters 9 and

10 include two from Stage 4 (Utilities, Transportation), three from Stage 3
(Retail, Insurance, and Marketing/Geodemographics), and three from Stage
2 (Agriculture, Banking/Finance, and Health Care). Additional industries in
Stages 3 and 4 are included in case studies throughout the book. The Stage
2 industries covered are rapidly growing and increasingly strategic. In this
and the next chapter, industries will be covered starting from the spatially
more mature ones in the upper right of Table 9.2 and working down and to
the left. A detailed case study of the large utility, Southern Company, appears
following the utility section. It sheds light on that industry and illustrates
organizational issues and ways to solve them.
UTILITIES
Summary: The major spatial applications are:

Energy procurement and planning
Electric and gas transmission
Electric and gas distribution
Customer care
Demand studies and economic development
Asset management
Supply chain and logistics
Finance and accounting
Corporate services, regulatory compliance
The Utilities industry in the United States is an old one and heavily reg-
ulated. Its challenge is to fight bureaucracy and keep up with rapid changes
on the outside. It is constrained from change by government regulations, the
long cycle times for approvals and building of utility infrastructure, and by
ingrained habits (Meehan, 2006). In spite of this, the industry often utilizes
advanced engineering and technology for the production, transmission, and
distribution of water, energy, and communications.
Among the shorter-term pressures and problems besetting utilities are
air pollution controls (a good thing for the public), blackouts such as the
Great Blackout of 2003, rise of consumer power to force changes, the Enron
energy pricing scandal and subsequent regulations, and the additional security,
regulation, and reporting in the outfall of 9/11 (Meehan, 2006). In response
to the problems, utilities have often put their focus on core competencies,
lowering costs, better customer service, more efficient operations, and use of
technologies to achieve improvements.
Electric and gas utilities produce energy in a process cycle portrayed
in Figure 9.1 (Meehan, 2006). Energy must be procured, transmitted, and
Figure 9.1 Business Model of Typical Electric or Gas Utility. Copyright © 2005
ESRI. All rights reserved. Used by permission
distributed. Corporate services, finance and accounting, and IT provide the

information and money flows for the physical processes. The supply chain
is essential for getting energy materials to the right places at the right time.
Monitoring and maintaining a “just-in-time” inventory-based supply chain
requires spatial technology since the inventory that is no longer in a warehouse
but moving around the world at large can be referenced with geospatial
coordinates. The business model for a communications utility looks similar
to this one, except procurement, transmission, and distribution are replaced
by sending, transmission, and receiving of communications.
GIS is a crucial technology at almost all points in the utility process cycle. In
energy procurement and planning, GIS is utilized to design and construct en-
ergy plant sites and transmission networks. If the power grid is not adequately
built, there are power losses or blackouts. An example is the design, construc-
tion, and continual monitoring of Great River Energy’s Pleasant Valley turbine
plant. GIS was utilized by engineers to visually analyze the plant design. Fol-
lowing the opening of the natural-gas-fired combustion plant in 2001, GIS
has been used to monitor plant component reliability. For example, in Figure
9.2 the number of failures in tiles in the turbine is shown visually for years
2001 to 2006 (Cox–Drake, 2006). In other cases, GIS can show the impacts of
power plant design alternatives on transmission lines, environment, surround-
ing neighborhoods and areas, and major customers. Outages in the utility grid
can be tracked down through examining outage reports and looking for com-
mon grid elements, which is assisted by GIS. For instance, for an electric
utility, all of the reported outages are on a given path and all have in common
a particular feeder line and two transformers. This diagnostic can direct repair
crews to the outage location more accurately. For electric utilities, the power
is transmitted on the electric power grid, which is owned by utilities. The
grid must be designed from a power engineering standpoint. GIS can produce
schematic drawings that show all the major network components and simu-
late loads, which are visualized symbolically. In this way, load scenarios can
be compared.
UTILITIES 269
Figure 9.2 Total Tile Failures in Unit-11 of Great River Energy’s Pleasant Val-
ley Natural-Gas-Fired Combustion Turbine Plant, 2001–2006. Source: Great River
Energy, 2007.
Other spatial concerns in transmission lines are to secure right-of-way, and

meet state and local regulations, while for built lines, the aim is to remain free
of vegetation intrusion or disaster damage. Many of these concerns are assisted
by study of the corridors surrounding the transmission lines. Spatial analysis
can place a buffer along the entire transmission grid, and phenomena within
the buffer can be studied, including land topography, vegetation, landholdings,
and security threats. Remote sensing can determine the type of vegetation
and hence if its anticipated growth is sufficient to pose a threat to the lines.
Satellite imagery and the grid can be overlaid, along with a buffer, as seen in
Figure 9.3, for a 400 kV line in rural areas of Great River Energy’s grid
in Minnesota, in order to evaluate and resolve the exposures and problems
within the buffer.
GIS is used to evaluate the encroachment of vegetation onto high-voltage
power lines. The vegetation threat to these lines becomes critical under cer-
tain conditions. In the New York Blackout that commenced on August 14,
2003 and affected more than fifty million people in the U.S. and Canada,
the primary cause was intrusion of trees onto transmission lines in Ohio
(Wingfield and Garnett, 2006). GIS is an excellent tool to assess the growth
and impingement of vegetation (Wingfield and Garnett, 2006). For exam-
ple, the New York Power Authority (NYPA) runs a vegetation management
program that checks on encroachment for 16,000 acres of rights-of-way
Figure 9.3 GIS and Satellite Overlay of 400 kV Transmission Line of Great River
Energy in Minnesota. Source: Great River Energy, 2007.
(ROW). The GIS application is linked to other systems that manage environ-
mental, land management, and equipment maintenance data. The GIS-ROW
application evaluates current vegetation and estimates the potential effective-
ness of vegetation control projects (Wingfield and Garnett, 2006). At NYPA,
GIS is set up as a portal, to provide diverse users with the work status of veg-
etation containment, as well as other operational and weather information.
GIS can be coordinated with utilities’ SCADA systems. SCADA (Su-
pervisory Control and Data Acquisition) senses, monitors, and controls the
chemical or physical processes of utility flows. For instance, SCADA for nat-
ural gas monitors regularly-spaced flow sensors that indicate the volume of
natural gas in the pipeline. Managers receive flow data for the whole network.
Spatial displays allow them to visualize and remedy problems occurring in
large-scale, complex networks. Some SCADA responses can be programmed
to automatically occur. Regional transmission lines can be monitored over
large geographies, even an entire nation. For example, a map of above normal
voltage readings for high-voltage regional transmission lines across the Mid-
Atlantic region, in Figure 9.4, can be studied to diagnose causes of problems
network-wide and alert local utilities to undertake maintenance or repairs.
Besides SCADA, GIS for utility distribution assists in outage manage-
ment, asset management, leak assessment, and “One Call.” Outages may be
UTILITIES 271
Figure 9.4 Regional Transmission Lines in Mid-Atlantic United States

(115–1000kV) Showing SCADA Above Normal Voltage Readings. Source: PJM
Interconnection.
due to equipment failures, high demand, aged or malfunctioning equipment,

or disasters. In an outage, utility customers face work stoppage, financial
loss, inconveniences, and safety risks. Using SCADA and other data, maps
can be produced of outage locations, both for internal response teams and
external community-based information. When Hurricane Isabel happened in
September of 2003 causing severe power losses in eastern Pennsylvania in its
aftermath (see Figure 9.5), GIS analysis revealed detailed patterns of outages,
which were useful to utility customer support, repair crews, and public safety
(PPL Utilities, 2007).
A common safety threat to a utility distribution network is that a building
contractor will accidentally cut a subterranean utility line, causing environ-
mental damage, economic loss, or injury (MapInfo, 2006). Most states require
a “One Call” system (sometimes known as “Call Before You Dig”), that con-
sists of an 800 number or website to inform a builder whether the subsurface
of a site is safe to go ahead with. Maps can pinpoint the underground util-
ity lines and store historical and descriptive attribute information. Figure 9.6
shows a “Call Before You Dig” GIS display screen for a water utility showing
housing parcels, subsurface water lines, and an incident record.
The field aspect of utilities concerns the operation, inspection, inventory,
maintenance, and replacement of the tens of thousands of physical facilities,
Figure 9.5 Outages in Eastern Pennsylvania from Hurricane Isabel. Source: PPL
Utilities.
Figure 9.6 Call Before You Dig Management for Water Utility. Source: Azteca
Systems. Copyright Azteca Systems, 2007.
UTILITIES 273
TABLE 9.3 The Leading GIS Applications for Electric and Gas Utility Firms, 2004
Leading GIS Applications for Electric Utility Firms

1. Trouble Call and Outage Analysis
2. Engineering Work Order Design
3. Field Automation and Workforce Automation
4. Work Management
5. Data Maintenance
6. Engineering Analysis
7. Customer Information System (CIS) Integration
8. Distribution Automation/SCADA Interface
9. Conversion and Data Capture
10. Executive Information Support System
The Leading GIS Applications for Gas Utility Firms

1. Mobile Collection, Viewing, and Access to Data
2. Work Management and Process Automation
3. CIS Integration with GIS and Facilities Management
4. Facility Maintenance
5. Management of Leak Detection
6. Asset Management and Plant Maintenance
7. Gas Outage Management System
8. Corrosion Protection
9. Replacement Maintenance
10. Marketing and Business Geographics
Reprinted with permission of the Geospatial Information & Technology Association, www.gita.org.
equipment, and devices. The significance of this aspect for GIS is seen in
Table 9.3 that lists leading GIS applications for electric and gas utility firms
in 2004 (GITA, 2005). Ones with major field components are mobile data
collection, managing work orders, leak detection, and outage management.
(GITA, 2005).
In the field, assets can be monitored and maintained through mobile, GPS-
based devices. Mobile spatial architectures, discussed in Chapter 5, are be-
coming more important. A utility’s web services can be extended to mobile
devices in the field (Figure 9.7). Although the screen size in the field is often
small as seen in the top of the figure, cartographic adjustments can be made
to simplify the symbol sets in the field, but keep them consistent with the
desktop (Rude, 2006). Map labeling may need to be done later, since crews in
the field are more keyed to their immediate surroundings than office workers
at desks, i.e. they’re more interested in plugging a pipeline leak than, docu-
menting what they are doing. However, redlining changes can be performed
in the field by working on a tablet PC as seen in the bottom of Figure 9.7
(City of Burbank, 2007).
Figure 9.7 Mobile Field Display of Neighborhood Utility Network. Sources:

Meehan, 2006; City of Burbank, 2007.
There are numerous benefits for a utility to make the transition to mobile
GIS for its field force. For instance, at Xcel Energy, Rude (2006) pointed to
benefits of greater efficiency, knowledge, and reduced travel (see Table 9.4).
The change to mobile can be challenging for field workers (Rude, 2006),
who are often loath to give up the daily paper maps they have used for years.
However, at Xcel Energy, crew members were optionally allowed to retain
paper maps in the transition, yet they all soon moved to a mobile GPS device,
as they perceived its greater knowledge base, functionality, and accuracy.
Utilities can provide better customer care by using GIS. Most utilities such
as electricity, gas, and water are metered. Meter reading has traditionally
been conducted by meter readers in trucks. Mapping displays in GPS-enabled
vehicles assist meter readers to locate customers. Managers plan and monitor
UTILITIES 275
TABLE 9.4 Benefits of Mobile Mapping, Based on Lessons Learned at Xcel Energy
1. Electronic gathering of large quantities of information in the field.
2. Graphic (map) display in field.
3. Drill down on field maps to detailed engineering data.
4. Variety of information in the field can be expanded, for instance gas leaks, reporting of
damaged facilities, and scouting of storm impacts.
5. Immediacy of input and ability to immediately notify crew mates.
6. Whole field team updated daily.
7. Crews can track distribution lines and pipes in much more detail. They can more
quickly and accurately locate values, switches, etc.
8. Reduced printing costs.
9. Broader variety of data accepted by field workers.
10. Reduced travel time and productivity improvements from mapping of work orders, for
best routing.
11. Use of GPS to identify assets in the field, to know where the field worker is, check on
driving directions, and communicate among the crew on locations.
Source: Rude, 2006. Reprinted with permission of the Geospatial Information & Technology Association,
www.gita.org.
locations of a whole force of meter readers. Some utilities have automated

meter reading. As the truck gets within Bluetooth-communications-range of
a house, the updated meter information is automatically transmitted to the
truck by Bluetooth, and then forwarded by wireless to a central database.
Other electrical utilities such as Southern California Edison with IntelliGrid,
offer more advanced two-way meters that send and receive information on an
hourly basis, monitoring home electrical devices and providing richer usage
data to the utility, as well as allowing the homeowner to better manage use
of utility devices (SCE, 2006). The utility has richer data to spatially model
household consumption and behavioral patterns.
Utilities can service customer inquiries through call centers and web cen-
ters that offer outage data, billing information, home energy management
tips, and service order tracking. GIS visualization assists service personnel
where location is pertinent such as for outage zones. In the case of emer-
gencies, utilities can expand call-center and web-based information flow to
customers. In normal situations, utilities conduct target marketing to bring in
new customers and retain existing ones.
At the regional level, utilities need to understand and forecast demand.
This involves detailed demographic and consumption modeling that includes
location. For instance, in the first decade of the twenty-first century, greater
Los Angeles metropolitan population is expanding rapidly to the east in
San Bernardino and Riverside counties. Utilities need to model population
growth, types of new businesses, and locations using demographic and GIS
software. Along with modeling, utilities support governments and businesses

in economic development, especially in fast-growing regions, such as the
aforementioned ones. For instance they participate in regional growth plan-
ning and contribute information and knowledge regarding the location of
present and future populations.
Finance and accounting were discussed in Chapter 4 in the context of ERP.
Since the location and proximities of utility assets are crucial to their valu-
ation and taxation, GIS systems track utilities’ extensive property and asset
holdings. This detailed finance and accounting information is also required
for certain regulatory filings such as for the Federal Energy Regulatory Com-
mission (FERC). It also becomes essential if the firms undergo merger or
acquisition.
Utility companies must comply with many government regulations, direc-
tives, and federal and state laws. One example is FERC’s Vegetation Manage-
ment Reporting Order that requires regular reports from power transmission
operators, controllers, and owners on current vegetation conditions (ESRI,
2004). Report information can be more efficiently collected by field crews
with GPS-enabled mobile devices, who can check compliance status on the
spot and take actions in the field to comply. The central database for reporting
can be updated in real-time as the field work gets completed.
A crucial gas-utility regulation is the management of pipeline integrity
in High Consequence Areas (HCAs). They are portions of the pipeline
where the flows and surroundings pose grave risks of damage and injury
if an incident occurred. Regulations require inspections, repairs, and mit-
igation steps. HCAs are examined through modeling of pipeline scenar-
ios. This is supported by GIS software with modeling extensions to assess
the HCA risks, perform spatial analysis of HCAs, and conduct associated
hydraulic simulations. In the field, a GPS-enabled mobile force can per-
form these tasks more efficiently. To recap, the vegetation and HCA exam-
ples illustrate the multiple spatial applications that respond to government
regulations.
Utilities are a prime concern for homeland security. The pulse beat of mod-
ern economies depends on reliable water, energy, and communications. As
seen in Table 9.5, threats specific to utilities include direct ones to facilities
and transmission lines, indirect targets that cause associated damages and in-
juries while reducing the ability to respond and recover, and other associated
facilities and events that may be impacted by loss of energy and communi-
cations (Kirk, 2006). In 9/11, electricity was cut off in parts of Manhattan.
Restoration and repairs were hampered by indirect target losses such as trans-
portation closings and hazardous fumes. The three-year-old New York City
Emergency Operations Center containing integrated GIS for the city located
in one of the twin towers was destroyed in the attacks. Within a day or two, a
new GIS center had to be rapidly set up and replacement data quickly gath-
ered from multiple sources that enabled mapping and analysis of outages and
utility lines (ESRI, 2002). Among the most important mapping contributions
UTILITIES 277
TABLE 9.5 GIS and Homeland Security: Threats to Utilities
Direct targets
Electric generation plants (nuclear, coal, gas, hydro)
Gas storage facilities
Water treatment plants
Oil refineries
Gas pipelines and pump stations
Electric transmission grid
Telecommunications central offices, switching stations, and substations
Utilities offices and other facilities
Indirect targets
Road closings and gas station losses
Other public transportation closings
Fires/explosions of non-utility assets nearby to damaged utility assets
Hazardous chemical releases
Disease outbreaks that lower workforce
Employee residential locations
Other associated facilities/events that could be targeted
Hospitals, universities, public buildings (indirect effect on utilities)
Transportation centers (indirect effect on utilities)
Historical sites
High-profile events (parades, rallies, elections, sporting or cultural events)
Source: Kirk, 2006. Reprinted with permission from the GITA 2006 Conference Proceedings, a publication
of the Geospatial Information & Technology Association, www.gita.org.
in the aftermath was to locate the oil fuel tanks and Freon tanks underneath
the World Trade Center site so that threat of explosion was averted.
Utilities can ready themselves for homeland security threats by target
identification, risk assessment, preparedness planning (pre-disaster), response
and recovery plans (post-disaster), and mitigation planning (post-recovery).
The steps can be consolidated into a comprehensive business continuity plan.
Geography is important. For instance, natural gas supply and transmission can
be protected ahead of time from major disruptions by mapping and analysis
of production facilities, networks, and storage units to highlight security
weaknesses and exposures, which can lead in turn to prevention by expanded
security surveillance and staffing (Kirk, 2006).
For any type of potential disaster, there is a need to share spatial information
beforehand with government agencies and industry partners (Kirk, 2006).
Shared information helps the partnered governments and businesses to prepare
together. Consequently, in a catastrophe the governments, utilities, and other
enterprises can more quickly assess the damage and injury impacts and know
the status of each others’ intact capabilities. Although utilities may resist the
cost of data sharing, the potential long-term benefits are reduced damages
and losses.
Enterprise utility applications are large-scale and can involve tens of thou-
sands of facilities and millions of customers. They may extend over multi-state
regions or entire nations. In Chapter 4, enterprise GIS was seen to be designed
to work together with other major functional systems such as marketing, sup-
ply chain, and finance/accounting. The Pidpa and Nesa case studies highlight
how enterprise integration can be achieved for a water and electrical utility.
Building a utility enterprise architecture involves workflow and process mod-
els, data models, databases, output design, underlying physical infrastructure,
integration with other corporate systems and with those of other organizations.
(Dangermond, 2006). An example is the Enmax case in Chapter 8. Another
example of a successful initiative to develop an enterprise utility system is
this chapter’s case of Southern Company. The practical implication is that
management needs to spend extra time considering the best enterprise design
incorporating utility processes and the underlying physical architecture, while
integrating the business and utility enterprise modules with a GIS.
CASE STUDY OF SOUTHERN COMPANY
Southern Company is one of the nation’s largest utility holding compa-

nies. Through its five operating companies, Georgia Power, Alabama Power,
Mississippi Power, Gulf Power, and Savannah Electric, the combined South-
ern Company serves 4.2 million customers in four states covering 120,000
square miles (Adams and Johnson, 2006). Prior to 2003, Southern had differ-
ent GIS implementations in each of the five operating companies and operated
them independently, with little inter-organizational coordination or data shar-
ing (ESRI, 2006). GIS applications included asset management, marketing,
SCADA, economic development support particularly by Georgia Power, real
estate especially at Alabama Power, and a limited mobile GIS-GPS system
for field workers.
In 2003, Southern Company’s management set up an Enterprise GIS Project
with the goal to consolidate investment in GIS into common architecture,
structure, and data across all operating units. Southern formed a GIS Core
Strategy team that reported to Southern’s IT steering committee. It consisted
of a technology management representative from each of the five operating
companies. In 2003 the team developed a strategic vision for spatial technolo-
gies across all companies and formalized an enterprise vendor alliance with
ESRI. A GIS management consultant was called in to formulate an approach
to implementing the enterprise plan. The architecture was client server with
a spatial database (ArcSDE) and specialized utility add-on software (ArcFM
from Miner&Miner).
The operating companies moved at different speeds to develop their mi-
gration to the enterprise standards. Alabama Power quickly implemented it
within a year so its results constituted a model for the others. By 2006, Gulf
CASE STUDY OF SOUTHERN COMPANY 279
Power, Mississippi Power, and Savannah Power had migrated to the common
architecture.
Georgia Power (GP) took a slower route and needed to call back the outside
GIS consultant to more deeply develop its plan to migrate from AutoCAD to
the new standard of ArcGIS/ArcSDE (ESRI, 2006). The Georgia Power im-
plementation project was finally approved in June 2005, with implementation
in 2007. Why was Georgia Power delayed? GP’s prior CAD system known
internally as Automated Mapping and Construction (AMC) was a very high
quality system that had good reviews from its over 300 users (Powers and
Johnson, 2006). It had taken considerable effort and investment to implement.
However, Southern’s needs trumped the local preference to keep AMC as is.
Yet the conversion of such a large existing system and user base would take
more time.
A GP project transition team was set up to do the conversion. It included
business and field utility employees, experienced systems analysts, ESRI
experts, and a consultant. The project team had to go slow as it worked
through the often contrasting opinions of strong-willed members (Powers
and Johnson, 2006). The team did not yield to “group think,” but went in the
opposite direction, often reviewing and reconsidering prior decisions, which
took more time (Powers and Johnson, 2006). The project was delayed but in
the end there was buy-off by the whole team of a conversion plan that was
approved and is now in implementation.
Overall, Southern considers that the biggest benefit of its enterprise GIS
is economic. Purchase in bulk lowers costs, and data and expertise can be
shared more efficiently among the units. A useful web-based application
TransView was developed in-house during the project to allow a common
interface for query and work design among the GIS teams in the operating
companies. (ESRI, 2006). Overall, the costs were high in terms of dollars
spent, people’s time especially at Georgia Power, and some intangible losses
of morale (hopefully temporary). Nevertheless, the change was strategic and
Southern expects to stay its course with this application for ten years or more
which provides ample time for payback. The Enterprise GIS Project correlates
with corporate strategy and was intentionally positioned high up to report to
the Corporate IT steering committee.
This case provides lessons regarding the organizational structure and lead-
ership of GIS for a large company. The project represented disruptive change.
Outside consulting was necessary to develop the unprecedented five-company
transitioning path. Expertise and motivation to make the change were more
available at some operating units than others. Georgia Power, the slowest to
transition, encountered some inertia and even resistance that reflected the large
success of its prior CAD implementation. Working through it organization-
ally required extended team discussions, but finally it gained consensus and
approval. The case highlights the importance for GIS success of leadership,
carefully planned change, consensus building, and overcoming resistance. It
also demonstrates the benefits of outside, uninterested expertise to gain clarity

during a period of organizational change.
TRANSPORTATION

Routing and scheduling
Vehicle tracking and dispatch
Emergency vehicle tracking
Fleet management
Aviation, airports, and maritime
Railroads
Spatial technologies are used extensively by businesses to plan, manage,

route, and maintain transport, and respond to emergencies. A key aspect is the
potential for real-time locational information exchange through combinations
of GIS, GPS, wireless, and satellite communications. For instance, Apex
Office Supply in Vinton, Iowa, is a small, twenty-year-old office supply firm
with 1,500 customers in a 50-mile surrounding region. Trucks are loaded in
the morning to deliver to retail outlets in the region. The daily loads need to
be balanced between drivers. The delivery person’s territory has been decided
beforehand through spatial modeling that balances territories throughout the
city for the delivery team. Balance refers to evenness among drivers in daily
time, workload, and costs.
GIS and optimization models can determine drivers’ routes for the day.
An on-board mobile computer provides route maps; database information on
load, customers, and schedule; real-time navigation; audio directions while
driving; the ability to re-route that day or the next if obstacles occur; and
capability to key in new locations and customers while in the field (ESRI,
2006).
Information is accumulated daily for the whole fleet and made available
to managers to model longer-term trends, and assess routing performance.
Another benefit is that the bar codes of items being loaded on a truck in
the morning can be checked against the driver’s manifest, and discrepancies
corrected. The driver has load sheets both in paper and on-line that list and
describe exactly the items for each stop. Although there was some driver
resistance, it was quelled by inviting drivers to attempt to “best the software”
with their own routing, but no driver was able to do this, and so the benefits
became more evident as a result (ESRI, 2006).
GIS routing and scheduling software is available from major software
vendors and specialists for industries. Vehicle tracking can be applied to entire
fleets of vehicles. For instance, a national trucking fleet utilizes GPS-based
TRANSPORTATION 281
Figure 9.8 Business Fleet Routing and Scheduling (ArcLogistcis Route from
ESRI). Copyright © 2006 ESRI. All rights reserved. Used by permission.
vehicle tracking to manage and monitor hundreds of large trucks nationwide.

This is referred to as automatic vehicle location (AVL), which is the capability
to determine the exact location of a vehicle and transmit the information in
real time to a centralized location for management use. Although locations
are usually registered by a GPS receiver on-board, inertial navigation or RFID
can also be used. AVL-managed fleets are present in trucking, construction,
business deliveries, police and fire, and public transport. AVL can be combined
with project and task management. For instance, as seen in Figure 9.8, a three-
truck fleet is being routed. At the lower right, a pert chart displays the day’s
time allocations for different color-coded tasks by each of the drivers. The
route for Truck 1 is shown on the screen. The driver has requested information
on the customer from the first stop of the day at 9:32 am arrival (ESRI, 2006).
The “order properties” window on the left provides information from the work
order for first stop, including service duration, items delivered, and the time-
delivery window, which was realized. Besides use to the driver, accumulated
daily information from such a system can be analyzed by the manager for
decision support on long-term fleet routing.
Sears, a case study examined in Chapter 7, uses AVL systems to support
its home service technicians who make fourteen million service calls yearly.
Delivery trucks, such as the one in Figure 9.9, have on board an AVL system
Figure 9.9 Sears Delivery Truck and Driver. Source: ESRI, 2006
that includes GPS, GIS, satellite and terrestrial communications, audio links,
and ruggedized laptops with touch screens (ESRI, 2006). Each truck has a
Mobile Base Station communications device on top that communicates infor-
mation every four seconds via wireless to the Sears Command Center. The
driver can view current location, his/her route, and route stops on the laptop
screen that is shown in Figure 9.10. The display functionality is simplified
to allow messaging, marketing information, an activity list, and other basic
features. If the driver’s route changes, a new one can be added by dynamic up-
dating that day. Each truck has the nationwide TeleAtlas street data on-board
(ESRI, 2006).
Emergency vehicle tracking follows a similar routing and AVL framework,
but has more need to optimize the fleet actions. For instance, a police depart-
ment has vehicle routing and AVL in a city. For a police emergency, the
dispatcher can see where the entire police fleet is located, which patrol cars
are nearest to the incident and what staff and equipment resources among
them are appropriate to deploy. Ultimately the dispatcher decides on which
vehicle(s) to send. Subsequently, the driver dispatched can monitor his/her
distance from the incident and other resources in case they are needed for
back-up. The difference for emergencies is the time urgency, the incident
coordination, and enhanced capability to optimize the fleet.
Aviation, airlines, and airports depend on GIS. Airline firms and air traffic
controllers plan routes in airspace through GIS. Map layers of satellite and
digital aerial photos are registered with those airport and flight-path features
to allow aircraft to be positioned exactly relative to other aircraft and ground
features. This can be useful for flight control, security considerations, noise
studies, and even planning of noise-impacted real estate developments. The
TRANSPORTATION 283
Figure 9.10 Sears Smart Toolbox Routing Feature. Source: ESRI, 2006
GIS imagery has become available in 3-D which especially helps assess
airspace availability as it gets more crowded (ESRI, 2006). A map of Las
Vegas International Airport and the surrounding city area overlaid with the
runway noise plumes, shown in Figure 9.11, allows business, government, and
community stakeholders to evaluate noise versus current and future building
developments in this rapidly growing city.
Another example of GIS for airports (Figure 9.12) is a mobile device for
monitoring terminal concessions relative to the airport layout. Within the
terminal, concessioned and unconcessioned spaces are colored differently to
allow tenants and airport management to understand the flows of customers
over time and space.
GIS is used extensively to assess security threats of airports and aircraft.
It “provides a powerful analytic capability for understanding vulnerability in
existing facilities as well as in pinpointing trends in incidents and past security
breaches” (ESRI, 2006). Another use that applies to airports, railway stations,
or other transport terminals is to display engineering design plans relative
to layers of transport, parking, hotels, commuter access, and neighborhood
proximities.
Railroads likewise are concerned about the geographic layout of their facil-
ities and networks. They need to know about and map terminal design, envi-
ronmental impacts, security, and the detailed locations of present or prospec-
tive maintenance problems. When tracks are out-of-use in maintenance,
Figure 9.11 Airport Runways with Noise Plumes. Source: McCarran International
Airport, 2007.
TRANSPORTATION 285
Figure 9.12 Mobile Device for Airport Inspection and Maintenance. Source: ESRI,
2006
“slow-order maps” are displayed that indicate locations of current track re-
pairs, when they will be finished, and the consequent slowing down of train
flows on the network. Some train companies such as Conrail have on-board,
real-time monitoring of engine performance (Lang, 2000). That information
can be viewed spatially for the entire locomotive fleet in a control center, with
color coding showing the severity of maintenance issues. This allows certain
Conrail engines to be diverted for maintenance, using spatial algorithms to
lower the cost to the nearest repair center (Lang, 2000).
The popular NavStar and OnStar devices in cars highlight the use of spatial
technologies for personal transport. These systems display road maps that are
complete for continental areas, and provide visual and audio directions to
find locations. The base maps are provided by companies such as TeleAtlas
that collect detailed and exact road-network information worldwide. Data are
available for the U.S., Canada, Europe, and increasingly in developing parts
of the world.
For advanced transportation problems, the modeling occurs at the level of

an SDSS or Spatial Business Intelligence. Algorithms from vehicle routing
models are combined with the spatial qualities of GIS (Keenan, 1998). For
instance, customized multi-vehicle, multi-depot routing models have been
studied by researchers for decades. The traditional operations research models
used simplistic representations of spatial relationships. The use of GIS offers
more sophisticated modeling of spatial relationships in routing. It also allows
interaction with other models and information in the GIS (Keenan, 1998).
Transportation is critical in logistics and supply chain applications, which
were discussed in Chapter 4. In summary, GIS and GPS have become essential
for modern transportation management whether by large transport firms,
smaller companies, or security and regulatory departments in business and
government.
RETAIL

Site location
Merchandising
Geodemographics, customer segmentation
Sales and marketing
Trade areas
Store renovation, cannibalization closure
The retail industry utilizes GIS with a focus on the customer who is key
to its sales and profits. Of particular interest are where the suitable customers
are located relative to the retail outlets, competitor outlets, transportation,
and other attractions and amenities. For larger retail chains, the balance of
national and regional products and markets is evaluated spatially. As seen
in Chapter 4, Customer Relationship Management Systems (CRMs) allow
retail firms to gather information on customers with emphasis on cultivating
and developing the best ones. The Chico’s case study illustrated how CRM
and GIS could form a strong pair of enterprise applications that succeeded in
enhancing Chico’s goal of achieving valued customer relationships. Chico’s
also used GIS for locating stores, direct mailings, and business intelligence.
This sub-section builds on that background.
Retail GIS accesses large variety and quantity of retail attribute data,
mostly governmental, some of which is presented in Table 9.6 (Roussel-
Dupre, 2002; ESRI, 2006). For spatial analysis, digital map layers of census
geography, business territories, and shopping center locations that correspond
to the attribute data are gathered and input into the GIS.
Site location is critical for any retail company. GIS allows mapping of the
sites of the firms’ retail outlets, that can be overlaid with sites of competing
RETAIL 287
TABLE 9.6 Retail Attribute Data Used for Spatial Applications in the Retail Industry
Demographic data
Decennial census data
One-year and five-year forecasts of census data
American Community Survey
Segmentation
Descriptions of the neighborhood profiles in the U.S. Example: Community Tapestry from
ESRI and PSYTE from MapInfo. Consists of 50–100 segments by using demographic and
socioeconomic composition.
Consumer expenditure
Consumer Expenditure Surveys from Bureau of Labor Statistics
Market potential
“Data on goods, services, attitudes, activities”
Projections of number of consumers. Gives indices for market potential.
Retail marketplace
Businesses can study retail supply and demand
Businesses
List of tens of millions of businesses giving “name, location, franchise code, industrial
classification code, number of employees, sales”
Shopping centers
Major databases on shopping malls. National Research Bureau and Directory of Major Malls.
Tens of thousands of shopping centers.
Traffic counts
U.S. national data on vehicular traffic
Modified from Roussel-Dupre, 2002, and ESRI, 2006
retail firms, as well as demographic, traffic, and neighborhood segmentation

(i.e. geodemographic) layers. Chico’s did this to locate new sites for its rapidly
growing branch network, to analyze cannibalization among its own stores,
and evaluate the locations of traffic and other access barriers. Cannibalization
refers reduction in revenues in old stores when a new one arrives.
Another example is Marco’s Pizza, a rapidly expanding chain founded
in Toledo, Ohio in 1978 (MapInfo, 2006). In 2005, Marco’s had 148 stores
and was Number 1 for growth among U.S. pizza chains. Marco’s performed
site location analysis by overlaying layers of its other and competitor lo-
cations, sales estimates, and the size of trade areas (see Figure 9.13). A
modeling software package calculated the prioritized best locations based
on the size and proximity to trade areas, buffer distances between sites and
other Marco’s/competitors’ locations, and distance from markets. As part of
this analysis, it used MapInfo’s PSYTE segmentation software to analyze the
neighborhood profiles of proposed locations and obtain forecasts of demand
(MapInfo, 2006). For a new site to be eligible, it had to have restaurant sales
potential of $550,000 or more yearly. The model showed the quantity, location,
Figure 9.13 Target Site for Marco’s Pizza in Las Vegas, with Marco’s and Com-
petitors’ Locations and Buffer Rings. Source: Marco’s Pizza, 2007
and priority of potential sites as well as each site’s index of trade-area sales
potential (MapInfo, 2006). Marco’s has used this spatial siting approach suc-
cessfully. Top management indicates it provides a “strategic blueprint for
expansion” (MapInfo, 2006).
Analysis of trade areas is important for deciding on locational sales ap-
proaches. A retail trade area is the geographic area from which the majority
of a store’s customers and/or sales originate. Mapping of trade areas is useful
because it reveals to a business what the competitive landscape is (Thrall and
Fandre, 2003). A number of models can be used to estimate it. The simplest
is a point map, i.e. one that shows where customers reside. A crude trade
area can be created around the points (Thrall and Fandre, 2003). A slightly
more sophisticated approach is radial analysis (also referred to as “ring anal-
ysis”), which assumes that a trade area is circular. Different diameters of rings
in the model account for certain percentages of a retailer’s customers. Levi
Strauss North America manufactures and markets clothing products, the most
famous of which are jeans. The firm and its distributors sought to analyze
for a hypothetical new retailer location what customers, existing authorized
retailers, and potential retailers were located in a trade area. A model of rings
around the potential location was created using the web-based ESRI Business
Analyst Online with 1-, 3-, and 5-mile radii. This was useful for decision
support on the new locations (ESRI, 2006). An example is seen in Figure
9.14 of such circles emanating from a hypothesized Levi Strauss location in
San Francisco.
More sophisticated trade area analysis can be done by the Huff Trade Area
Model (Dramowicz, 2005). This model formulated by David Huff (1963)
RETAIL 289
Figure 9.14 Web-Based Ring Analysis for Prospective Retailers.
maps a probability surface for customers that is the ratio of a measure of

attractiveness of a particular store (or shopping center) divided by a distance
measure, to the sum of attractiveness of all stores (or shopping centers)
divided by a distance measure. Hence this trade area measure not only takes
into account distance, but also the attractiveness of stores and the influence
of competitors.
The Huff trade-area model is formally represented as follows:
−␤
A␣j Di j
Pi j =

n
−␤
A␣j Di j
j=1
Where A j is a measure of attractiveness of store j, such as square footage,

Di j is the distance from i to j,
␣ is the attractiveness parameter estimated from empirical observa-
tions,
␤ is the distance decay parameter estimated from empirical observa-
tions,
n is the number of stores including store j.
Figure 9.15 Application of Huff’s Model to Estimate the Trade Area for Micmac
Mall, Dartmouth, Nova Scotia. Note: attractiveness of a shopping mall was deter-
mined by a single variable, Gross Leasable Area. Source: Dramowicz, 2005
−␤
The term A␣j Di j is called the perceived utility of store j by a consumer
located at i.
The ␣ parameter is the exponent to which store attractiveness is raised. It
allows a nonlinear attractiveness.
The ␤ parameter simulates the decay rate of the drawing power of the store
depending on customer’s distance from it. If ␤ is higher, there would be less
influence of the store on more distant customers.
The model is illustrated in Figure 9.15 for the Micmac Mall in Dartmouth,
Nova Scotia, where the store attractiveness is Gross Leasable Area (Dramow-
icz, 2005). It’s evident that Micmac Mall’s trade area forms an oblong shape
that is circular towards the competitors but bulges out in the northeast direc-
tion, away from the competitors. This information can be useful in under-
standing locational market draw. It informs marketing campaigns tailored to
different zones of the trade area. A weakness of Huff’s model is that data on
attractiveness and the locations of all the competing stores must be gathered
and updated.
Another retail use of spatial analysis is to assess spatial aspects of national
marketing initiatives. For large retail firms, products and marketing programs
RETAIL 291
are brought forward on a national basis. However, the local customers, com-
petitors, tastes, markets, and environments differ around the country. Spa-
tial analysis can be applied to make comparisons, model geodemographic
influences, assess consumer tastes, and look more broadly at geographic
strategies to engage nationally against competitors, locate different sizes of
stores optimally, avoid cannibalization, perform well-reasoned closures and
relocations, assess demand, and take into account regional population and
economic growth. These efforts require sophisticated models and lots of
data, resources, and oftentimes consulting expertise. The benefits, however,
are to sharpen the national thrust of marketing, rising above local specific
issues.
Jo-Ann Stores provides an example of such broader retail marketing. The
firm, founded in 1943, is the leading fabric retailer in the United States. Par-
ticularly for families or individuals engaged in hobbies and crafts, the firm
is dominant, capturing about three-quarters of the household market. The
products consist of crafts, framing and floral products, fabrics for home dec-
orating, seasonal home items, and varied home accessories. In 2005, Jo-Ann
Stores had 915 traditional stores in forty-nine states and seventy superstores
in sixteen states. Superstores first appeared in 1995. The map of Jo-Ann
Stores in New York and surrounding states in Figure 9.16 reveals nine su-
perstores and thirty-three traditional ones for New York state (Jo-Ann Stores,
2006).
Five years after the superstores were rolled out, the marketing problem
confronting Jo-Ann was to develop a future plan for the superstores, partic-
ularly what mix of superstores to have, whether superstores would attract a
different demographic profile than traditional ones, and where to locate new
superstores (MapInfo, 2006). Jo-Ann first used its three-million record cus-
tomer base to compare the customer demographics of the two store types,
and it found to its surprise that they were the same regardless of what region
around the country. The analysis implied that a group of traditional stores
could be relocated and combined into a superstore without losing customers
while achieving economic efficiencies. Subsequently, new superstores were
added without worrying about losing prior customers from the traditional
ones facing closure (MapInfo, 2006).
Retail companies are heavy users of GIS for marketing, site locations,
CRM, and in-store merchandising. They are able to gain enhanced under-
standing of the geographies of customers, competitors, products, store types,
and geodemographics leading to better performance and market appeal. GIS
often leads to improved integration across corporate databases (Roussel-
Dupre, 2002). As seen for the Chico’s case, some of the spatial techniques
are sophisticated and the databases can be large, but firms such as Chico’s,
Marco’s, Levi Strauss, and Jo-Ann have planned the technology carefully
and understood the necessary investment and training, leading to competitive
gains.
Figure 9.16 Jo-Ann Stores’ Web-Based Customer Application to Access Retail

Outlet Information. Source: Jo-Ann Stores, 2006
CHAPTER SUMMARY
Spatial technologies are implemented in organizations. Spatial technologies

can alter the organization and vice versa. GIS tends to be the responsibility
of individuals in small businesses and specialized GIS departments in middle
and larger ones. Since IT is usually a separate and larger unit, a compatible
organizational relationship between GIS and IT is important for success. The
Southern Company case illustrates organizational change accompanying the
roll-out of spatial enterprise GIS, consolidating five prior GIS applications for
separate units. Some organizational resistance was encountered but worked
through.
The chapter applies Nolan stage theory to categorize twenty-five industries
by their spatial maturity. There is a range from initiation through integration
stages. This chapter discusses three of the most mature and important indus-
tries for GIS, Utilities, Transportation, and Retail. The next chapter examines
five industries in the contagion and control stages for GIS. The goal of pre-
senting the industry applications is to emphasize that GIS has significant
REFERENCES 293
applications specific to users and customers in particular industries. The busi-

ness person in GIS needs to be aware of these distinctions and know the most
important ones. The Utilities industry is naturally spatial and has urgent real-
time applications including service for the customers and field-work by the
company, which is becoming mobile-GPS enabled. Transportation has been
revolutionized by locational technologies that allow improved navigation of
the vehicles while giving management better tools for routing, maintenance,
and security. Retail can strengthen its marketing and customer relationships
by combining GIS with CRM, siting analysis, geogdemographics, and with
marketing methods such as direct mailing. Singly or in combination, these
methods sharpen locational decision-making in retail and increase competi-
tiveness and customer awareness, care, and services.
REFERENCES
Clarke, Ian, and Jennifer Rowley. 1995. “A Case for Spatial Decision-Support Sys-
tems in Retail Location Planning.” International Journal of Retail and Distribution
Management, 23 (3): 4–10.
2006 Proceedings.
Daratech. 2004. Geographic Information Systems Markets and Opportunities. Cam-
bridge, MA: Daratech.
Directions Magazine. 2002. “Navigation Technologies Unveils Real-Time Traffic So-
lution.” Directions Magazine, October 14. Available at www.directionsmag.com.
Directions Magazine. 2004. “NAVTEQ Announces TrafficData Solution for North
America.” Directions Magazine, April 8. Available at www.directionsmag.com.
Dramowicz, Ela. 2005. “Retail Trade Area Analysis Using the Huff Model.” Direc-
tions Magazine, October 19.
ESRI 2002. “New York City—Creating a Disaster Management GIS on the Fly.”
ArcNews, Winter. Available at www.esri.com.
ESRI 2004. “GIS Solutions for Regulatory Compliance.” Energy Currents, Fall,
Redlands, CA: ESRI Inc., pp. 1–6.
ESRI 2006. “GIS Solutions for Retail.” Redlands, CA: ESRI Inc.
Fritz, Oliver, and Petter Skerfving. 2005. “Monitoring and Analysis of Power Line
Failures: An Example of the Role of GIS.” In Pick, James B., Geographic In-
formation Systems in Business, Hershey, PA: Idea Group Publishing, pp. 301–
323.
Gaudet, Cyndi, Helen Annulis, and John Carr. 2001. “Workforce Development
Models for Geospatial Technology.” GeoSpatial Workforce Development Cen-
ter, The University of Southern Mississippi, Hattiesburg, MS. Available at
www.geowdc.com/research/research.htm.
Geotab. “Welcome to the Geotab Checkmate Features Section.” Available at
www.geotab.com.
GeoWorld. 2005. “Industry Outlook 2005.” GeoWorld, December, pp. 24–31.
GITA. 2005. “Geospatial Technology Survey.” Aurora, CO: Geospatial Information

and Technology Association.
GPS World. 2004. “NAVTEQ.XM satellite hit the road.” GPS World, May. Available
at http://www.findarticles.com.
Kirk, Kevin. 2006. “The Need for Geographic Information, Analysis, and Sharing
for Homeland Security in the Utilities Industry.” Proceedings of 2006 GITA
Conference.
MapInfo Inc. 2006. “PCOS: Web-enabled ‘Call Before You Dig’ Application Cuts
Operating Costs for Utilities and Provides Greater Public Safety.” Troy, NY:
MapInfo Corporation. Available at www.mapinfo.com, September 1.
Marble, Duane F. 2006. “Who are We? Defining the Geospatial Workforce.” Geospa-
tial Solutions, May, pp. 14–21.
Meehan, William J. 2006. “Enterprise GIS: If You Build It, It Will Fund.” GeoWorld,
March.
Pun-Cheng, Lillian S.C., and Alice W.C. Chu. 2005. “GIS in Business—Routing
Analysis of Plaza Shoppers.” Proceedings of Map Asia 2004, Beijing, China,
August 26–29.
Ray, Julian J. 2005. “A Web-based Spatial Decision Support System Optimizes
Routes for Oversize/overweight Vehicles in Delaware.” Decision Support Systems,
in press.
Rogers, E.M. 1995. Diffusion of Innovations, 4th Edition. New York: Free Press.
Roussel-Dupre, Stephanie. 2002. “What Are Your Customers Telling You?” Inte-
grated Solutions for Retailers. Available at www.ismretail.com.
Rude, Tony. 2006. “Moving GIS from the Office to the Field.” GITA 2006 Conference
Proceedings.
SCE. 2006. “Designing the Future: A Smart Grid Newsletter Case Study.” Rosemead,
CA: Southern California Edison, November, 11 pp.
Segal, Donald B. 1998. “Retail Trade Area Analysis: Concepts and New Approaches.”
Directions Magazine, November 18. Available at www.directionsmag.com.
CHAPTER 10
INDUSTRIES AND OUTSOURCING
INTRODUCTION
In this chapter, the applications of GIS in industries are explored further. The
chapter follows the industry maturity-stage framework introduced in Chapter
9. The objective is to explain specialized spatial applications in important
GIS sectors. The emphasis is on the distinctive aspects of these sectoral
applications, rather than on commonalities which have been the focus of
earlier chapters. Five key industries in GIS maturity stages 2 (Contagion) and 3
(Control) are examined, insurance, marketing/geodemographics, banking and
finance, agriculture, and health care. The chapter provides breadth of insight
into the benefits, costs, and challenges of these particular industries, illustrated
by case studies. The chapter concludes with a discussion of GIS outsourcing.
The chapter begins by discussing with two industries classified in Table
9.2 in Stage 3 (Control), namely Insurance and Marketing/Geodemographics.
Insurance utilizes GIS both externally with customers and internally to run
complex insurance operations, while Marketing/Geodemographics, uses GIS
for geo-segmentation, target marketing, CRM, direct mailing, and global
marketing. Three industries classified in Stage 2 (Contagion) are discussed:
Banking and Finance, Agriculture, and Health Care. In the former, GIS is
applied to site location, branch networks, trade area analysis, marketing, and
portfolio analysis. The reinsurance firm SwissRe has vast spatial capabilities
especially for disasters. The Agriculture industry has changed rapidly due to
precision agriculture i.e. applying agricultural cultivation and treatments to
tiny micro-plots, based on a combination of GPS, environmental sensors, GIS,
295
296 INDUSTRIES AND OUTSOURCING
and mobile technology. The spatial information can be stored and analyzed
to aid farming decisions and strategies. Health Care has diverse uses of
GIS, that range from epidemiological study to locational analysis of patients,
physicians, and facilities; emergency dispatch of ambulances; mapping of
hospital beds; and target marketing of drugs to physicians.
Further insight is provided by case studies. In banking, a Large Credit Bank
is challenged in breaking out of its traditional “legacy” orientation by demands
for use of new technologies including GIS in the branch networks of acquired
banks. For agriculture, Cargill’s Palm Oil operations shows how GIS can help
to plan plantations. In Health Care, a small firm, MapGistics, provides spatial
analysis of bed occupancies in hospitals, while the larger Kaiser Permanente
has multiple GIS uses in site locations, patient and member travel, and much
more.
The final section of the chapter considers the growing trend towards GIS
outsourcing, which can be local, regional, or global. In IT, outsourcing has
become prevalent and continues to grow. It has tended to start with outsourcing
of low-level IT functions such as data entry and move to higher ones such
as systems planning and designing database architecture. Another trend has
been the growing proportion of global outsourcing, i.e. to outsourcers in
another country such as India. GIS outsourcing resembles IT outsourcing
in most respects but has some differences. The case of Southern California
Edison’s GIS contract with Patni illustrates the benefits and challenges of
global outsourcing.
INSURANCE

Risk analysis
Catastrophe management
Business continuity planning
Claims
Customer analytics
Support for agents
Marketing
Insurance companies inherently are concerned with risk. Specifically, un-

derwriting is the process of calculating the rates and pricing for insurance
premiums incorporating risk analysis. GIS helps develop the process by mea-
suring proximities to risk. For casualty and property insurance, GIS can over-
lay risk exposures to natural hazards such as hurricanes and earthquakes with
man-made ones such as oil spills and terrorism events. GIS helps improve
the statistical confidence in risk estimates, by taking into account location.
INSURANCE 297
It was seen in Chapter 7 that Norwich Union, the leading British insurance
firm, produced a detailed flood map of the UK as an early and successful
step into GIS. Presently, its GIS for floods includes the analysis of elevations
of individual residential and commercial properties throughout the UK to
evaluate risk. Another example, RVOS Insurance, a small property insurance
firm in Texas, uses GIS to map risk concentrations (MapInfo, 2006). The firm
checks for extent of locational diversification of risk, and make changes over
time to diversify risk more.
In Chapter 3, a prototype spatial decision support system for assessing the
risk of typhoons in China was examined. It utilizes expert systems, spatial
statistics, and graphical displays to analyze historical data on typhoons in
order to model the pricing of premiums.
Besides identifying risk zones and mapping estimated premiums, there is
further spatial modeling and analysis that can be done. These include: (1)
developing models to diversify geographically the exposures for high-risk
policies, (2) developing region-specific models to determine risk and pricing,
and (3) projecting the types of claims that are likely for a region (ESRI, 2006).
The model outputs lead to preventative steps for insurers so policy holders
can reduce risk and obtain a lower rate (ESRI, 2006).
Since inevitably catastrophes will occur, including hurricanes, tsunamis,
or terrorism, insurance firms must provide support and assistance to poli-
cyholders who are impacted with losses. They need to be able to spatially
forecast unfolding events and consolidate a total picture of losses. The Large
Insurance Company case study in Chapter 3 concerned catastrophe analysis
and management. It demonstrated the use of GIS to avoid massing too many
policies in the same location.
Another service that insurance firms provide to clients is Business Conti-
nuity Planning (BCP). BCP consists of the policies and procedures to keep
a firm’s key systems running in the event of a disaster. A well-thought-out
BCP makes it easier for an insurance company to respond to and offer help
in a catastrophe. Among the contributions of GIS to BCP are the following
(ESRI, 2006):
r Reinforces the capability for emergency response

r Helps a firm in a disaster make necessary notifications
r Helps manage the gathering of field data in a disaster
r Helps manage and account for corporate assets
r Improves the preventative sharing of data between the insured and gov-
ernment agencies
r Adds maps to the BCP itself to understand the spatial dimensions of
business continuity processes
An insurance claimant is often distressed and seeks quick, even immediate

service. The provision of service to the client is however complex, involving
multiple locations, such as those of the incident, claimant, claims adjuster,

and insurance claim offices. The company must decide how to allocate
the adjusters. An SDSS that considers the workload and capability of ad-
justers and their driving distances from the incident or claimant can optimize
the allocation of claims adjusters.
SDSS is also used to analyze service areas, assess historical adjuster cov-
erage, and make what-if assumptions on adjuster numbers, territories, spe-
cialties, and schedules. This feature is available in Oracle’s Siebel Systems
Insurance Module, which can be connected to MapInfo software to check
on the proximities of policyholders, commercial clients, dealers, agents, and
sales and service territories (MapInfo, 2005).
GIS helps in monitoring compliance with federal insurance regulations and
ratings. An example is the Florida Farm Bureau Insurance Companies (FFB).
FFB provides auto, farm, homeowners, dwelling-fire, and business-owner in-
surance to members of the Florida Farm Bureau Federation. In 2004, FFB
insured $23 billion in property. GIS supports regulatory applications at FFB
through GeoVerify, a customized integration of data on regulations, rating ter-
ritories, and policyholder locations (MapInfo, 2006). The maps were overlaid,
so discrepancies between policyholders’ regulation ratings and those actually
applied in the customers records could be compared, and corrected in the
records at the next renewal time. FFB’s ROI for GeoVerify in the first year was
1000 percent (MapInfo, 2006). Generally, any regulatory requirement that has
locational differences is opportune for GIS. A related application is to check
policy premiums by location in tax jurisdictions and make record corrections
whenever a customer’s record shows incorrect property taxes. Property taxes
serve as one indicator of property value and attendant risk (Bedell, 2006).
Insurance firms compete to acquire customers and then the firms must
continue to provide care and services to retain them. To accomplish this,
marketing strategy includes spatial methods such as tailoring the locations of
marketing campaigns to particular customer profiles, focusing the sites for
direct mailings, and enabling call centers to map caller locations. GIS can
contribute to target marketing, geodemographics, CRM, customer service,
and marketing campaigns. Several variables specific to insurance customers
that are important to map in marketing studies are the premium rate structures
by region and customers by categories of insurance risk.
Insurance agents also benefit by GIS. Maps of sales territories by type of
policy can be used to determine workload balance among agent teams. GIS
can help the human resources department in mapping recruitment pools of
agents (ESRI, 2006).
SWISS REINSURANCE COMPANY
Swiss Reinsurance Company (SwissRe) is a 145-year-old reinsurance com-

pany headquartered in Zurich (Figure 10.1). Reinsurance refers to insurance
SWISS REINSURANCE COMPANY 299
Figure 10.1 SwissRe Headquarters Building in Zurich, Switzerland. Source:

SwissRe, 2006
that shares risk beyond what direct insurance firms can carry by themselves.
SwissRe is the world’s largest reinsurance company with 2005 sales of $26.6
billion (Hoovers, 2006). It offers health, property, life, auto, and liability rein-
surance mostly to small- and medium-sized insurers that cannot afford the
full risk of certain coverages.
At the core of its business is the GEOdatabase (see Figure 10.2), a vast com-
pilation of business data from private insurance data firms such as GFK Ma-
con, map data provider Europa Technologies, U.S. Census for urban sprawls,
ESRI for topographic datasets, CRESTA for insurance industry zones, and en-
vironmental data providers. GEOdatabase is connected to other systems such
as Radio Data System (RDS) for navigation and location-based solutions,
data of International Hydropower Association (IHA), SICS/nt (reinsurance
administration software), the Incremental Update File (IUF) which receives
datasets and updates GEOdatabase on a time-critical basis, and CatNet, the
Figure 10.2 Available Data in SwissRe’s GEOdatabase. Source: Bayerl, 2006
firm’s own global natural hazards information and mapping system (Bayerl,
2006).
This case focuses on CatNet (Catastrophe Network) which has been avail-
able to SwissRe and its direct insurers since 2000. It is a web services appli-
cation, based on ESRI’s ArcIMS server and ArcSDE data for management,
and the Oracle database. The web design is Java based (J2EE and Java ob-
jects), with the XML data exchange standard. The architecture is a form of
the web-based spatial applications covered in Chapter 5. Web services have
been slower to catch on in insurance so CatNet stands out as an early and
sophisticated example.
CatNet provides a detailed multi-layer view of major natural perils for al-
most any location worldwide. In fact a user can search for 650,000 locations,
request estimated risk rates, show all potential natural hazards for a loca-
tion, and see chronological tracking of past catastrophic sequences (SwissRe,
2006; Bayerl, 2006). An example of an application is mapping historical
disaster events and intensities for “earthquakes and tropical cyclones world-
wide, floods in selected countries, windstorms in Europe and tornadoes in the
U.S.” (SwissRe, 2006). The number of natural hazards stored is considerable,
including 2,500 cyclones for the twentieth century, and events on 1,500 vol-
canoes. Within minutes, a user can accurately visualize hazards and estimate
risks for any place on earth (Schmidt, 2002).
In the tracking of historical hurricane paths in the U.S. CatNet can access
data from the GEOdatabase on urban sprawl, places/locations, administrative
boundaries, topographic data, hazard data, and CRESTA zones for insurance
risk exposures. Figure 10.3 is an example of a user request that shows the
paths of devastating hurricanes Wilma, Katrina, and Rita overlaid on layers
showing cities, urban areas, capitals, roads, rivers/canals, and water bodies.
SWISS REINSURANCE COMPANY 301
Figure 10.3 CatNet—Tracking of U.S. Hurricane Paths. Source: SwissRe, 2006
Although not all of this appears in the figure, users can drill down to get more
specific information and map greater detail. In the case of Hurricane Katrina,
SwissRe and some its direct insurers were responsible for claims stemming
from damages in six states (Table 10.1) such as structural damage (Figure
10.4). CatNet supported rapid, interactive access to information on Katrina
to the SwissRe and direct insurers during and after the event. It now has
available the historical information and maps on Katrina and other hurricanes
to help insurers and insured plan for risks and exposures in the Gulf of Mexico
region.
TABLE 10.1 Losses and Insurance Claims from Hurricane Katrina, by state
State Losses ($millions) No. of Claims
Louisiana 22,600 900,000

Mississippi 9,800 490,000
Alabama 1,300 123,000
Florida 458 110,000
Tennessee 46 8,400
Georgia 22 3,300
TOTAL 34,226 1,634,700
(Source: preliminary estimates by ISO’s Property Claim Services Unit, 10/2005)

Figure 10.4 Structural Damage to Superdome in New Orleans from Hurricane

Katrina. Source: FEMA, 2005
MARKETING/GEODEMOGRAPHICS

Target marketing
CRM
Geodemographics
Spatial adoption/diffusion
The goals of Marketing and Geodemographics are to analyze the markets

for products and services and achieve success in marketing to potential and
existing customers. Marketing and geodemographics exist not only as a sep-
arate industry, but also exist within firms across a spectrum of industries. For
the retail industry covered in Chapter 9 and the insurance industry in this
chapter, most companies have internal departments that perform marketing
and geodemographics.
The role of GIS in marketing is to track and analyze the map layers of
locations of customers, employees, products, and stores, overlaid with layers
of transportation networks, topography, geodemographic profiles, and other
layers such as financial, economic, and behavioral. This section goes over the
main marketing/geodemographic spatial applications including target market-
ing, market segmentation, customer relationship management, and analysis
of diffusion. A case study is presented that applies spatial adoption/diffusion
MARKETING/GEODEMOGRAPHICS 303
methods to the roll out of a loyalty adoption program for a large retailer in a
major metropolitan area (Allaway et al., 2003, 2005).
Target marketing is the process of searching for, identifying, acquiring, and
retaining customers. Simple target marketing examines potential customers
based on a single attribute to identify those with greater potential (Bourgault,
2005). Spatial analysis can display customer density and dispersion relative
to siting of customer facilities (Bourgault, 2005). For instance, a national
fast-food chain can map for a region the density of customers’ residences
relative to three fast-food outlets of one brand in the area and five competitor
outlets. A further analysis leads to maps of the travel distances for customers
to each of the outlet locations. Regional management can better decide how
to add additional outlets and whether to close or re-locate existing locations.
Several techniques that can include GIS are helpful in target marketing
and other types of marketing. Statistical analysis can project attributes or
groups of customers or outlets based on particular characteristics, and the
results can be mapped. Regressions forecast one dependent attribute, say store
spending, based on historical data on other attributes which are considered to
be associated with it. Spatial regression is also available that adjusts for spatial
differences in forecasting the dependent variable. For instance, based on
averages for a ZIP code, over a thirty-year period, customers’ store spending
on mufflers (dependent attribute) is associated through regression with percent
of car ownership, income, and average commuting time. After estimating the
next five years of values for the three independent variables, store spending
on mufflers can be projected for five years. The forecast can be done for all the
ZIP codes in a metropolitan area, and a map created of the projected values for
muffler spending. The map is helpful in target marketing for muffler buyers.
In addition to regression techniques, CHAID (Chi-Square Automatic In-
teraction Detector) can also be used for forecasting. CHAID is an exploratory
method for forecasting where a qualitative or quantitative dependent variable
can be projected by a complex model that chooses the predictor variables and
specifies interactions among them in order to optimally predict the dependent
variable. The predictor variable could be a target marketing variable of inter-
est to a business. CHAID can be run for a group of areas, such as the ZIP
codes in a county or the cities in a large metropolitan area, and the estimated
dependent variable can be mapped.
CRM (Customer Relationship Management), covered in Chapter 4, is a
business application or strategy that is intended to enhance customer sat-
isfaction and expand revenues and profits. For larger firms, CRM may be
implemented as an enterprise system that follows customers intensively for
long periods of time and provides personalized services to help develop long
term relationships between customer and marketer. GIS can be connected to
CRM so customers’ locations are taken into account in providing the services.
The Chico’s case study illustrated how CRM and GIS complement each other
so the rich information from a CRM can be strengthened through knowledge
of location, and vice versa.
Geodemographics is a segment of the Marketing Industry defined as anal-

ysis of people by the locations in which they live (Sleight, 1997). Geode-
mographics helps identify market segments and consequently enables the
appropriate targeting of markets. If you live in an area, geodemographics
assumes you are likely to have the characteristics and behaviors associated
with that zone. For instance, if you live in a neighborhood profiled as “Pros-
perous Empty Nesters,” geodemographics assumes you are likely to resemble
that profile. Of course, there are errors built into the assumptions, e.g. you
might be of the profile “Aspiring Young Family” living in an “Empty Nester”
neighborhood. However, geodemographics is willing to tolerate some errors,
as long as the neighborhood is correctly assigned for most of the residents.
Geodemographics is significant in size: one study estimated its revenues
for 2000 at $300 million (Weiss, 2000). Geodemographics applications in
marketing, in addition to segmentation, include survey design, retail planning,
direct marketing, advertising, and marketing media analysis (Harris et al.,
2004). The advantage of geodemographics is that once the small areas of
a region have been characterized into segments, then a person living in a
particular small area can be ascribed a set of characteristics without having
to tediously research that person’s actual characteristics. Some functions of
GIS for geodemographics are the following (Harris et al., 2004): (1) connect
survey results to general-purpose characteristics of the geographical zones of
the region, (2) group the data based on geodemographic zones, (3) perform
statistical studies for a larger set of identical geodemographic zones, and (4)
visualize the results of geodemographic analysis in maps, charts, and graphs.
For instance, for (1), a marketing survey is conducted over a zone of twenty-
five ZIP codes to determine the extent of appeal of Magazine X. The appeal
can then be compared to the segmentation profiles for the twenty-five zones.
Community Tapestry, a commercial geodemographics product for the
United States, has fifty-five segmentation categories (ESRI, 2006). They are
grouped into the twelve LifeMode Summary Groups and shown in Table 10.2.
Figure 10.5 gives the income ranges for the twelve groups. The income ranges
are useful in identifying potential buyers for a product or service. The national
distribution of LifeMode categories for the U.S. is mapped in Figure 10.6.
There are dramatic differences between regions. For instance, the High
Society LifeMode covers seven categories and is defined as follows: “afflu-
ent, well-educated, married-couple homeowners” (ESRI, 2006). It clusters in
affluent metropolitan counties nationwide, including ones on the West Coast
of San Diego, Santa Barbara, San Francisco and the Silicon Valley, Marin
County, Portland, and Seattle. The Coastal Northeast falls into this category,
stretching from northern Delaware though Philadelphia, New Jersey, New
York City and into Connecticut, Rhode Island up to the Boston Area, as
do the major metropolitan areas in the Midwest. Another way to divide up
the fifty-five categories is into a dozen Urbanization Summary Groups (ESRI,
2006). The groups fall along an urban-rural range that includes such categories
TABLE 10.2 Community Tapestry Categories, Arranged by LifeMode Groups
High Society Senior Sun Seekers

Top Rung Heartland Communities
Suburban Splendor Scholars and Patriots
Connoisseurs Military Proximity
Boomburgs College Towns
Wealthy Seaboard Suburbs Dorms to Diplomas
Sophisticated Squires
High Hopes
Exurbanites
Aspiring Young Families
Upscale Avenues Great Expectations
Urban Chic
Global Roots
Pleasant-Ville
International Marketplace
Pacific Heights
Industrious Urban Fringe
In Style
Urban Melting Pot
Enterprising Professionals
Green Acres Family Portrait
Up and Coming Families
Metropolis
Milk and Cookies
City Lights
Urban Villages
Metropolitans
City Strivers Traditional Living
Metro City Edge Main Street USA
Urban Rows Rustbelt Transitions
Modest Income Homes Midlife Junction
Family Foundations
Solo Acts
Laptops and Lattes Factories and Farms
Trendsetters Salt of the Earth
Metro Renters Prairie Living
Old and Newcomers Southern Satellites
Young and Restless Home Town
Rural Bypasses
Senior Styles
Prosperous Empty Nesters American Quilt
Silver and Gold Midland Crowd
Rustbelt Retirees Rural Resort Dwellers
Retirement Communities Crossroads
The Elders Rooted Rural
Copyright © 2006 ESRI. All rights reserved. Used by permission.
as principal urban centers, urban outskirts, and small towns. Other companies
offer different geodemographic segmentations, such as the seventy-two cate-
gories in MapInfo’s PSYTE, which are available for 208,000 neighborhoods
in the U.S., as well as for Canada (MapInfo, 2006). The PSYTE categories
were constructed from block-group level census data through a combination
of principal components analysis and cluster analysis (MapInfo, 2006).
Figure 10.5 Community Tapestry Income Range for LifeMode Groups. Copyright
© 2006 ESRI. All rights reserved. Used by permission
A spatial example of applying PSYTE is Ekornes USA, the U.S. subsidiary

of furniture manufacturer Ekornes, headquartered in Norway. Ekornes USA
had problems in evaluating the performance of its U.S. dealers, since it didn’t
know the locations of its best potential customers. Using PSTYE, it identified
a priority target market segment as follows: “Couples between the ages of
45–65 whose children no longer live at home and who have a median fam-
ily income of $75,000 and above. They lead an active, affluent lifestyle and
vacation often, especially abroad. They go to the theatre and read more than
the average person. As an age group, they tend to act and feel as through
they are 10 years younger than they are.” (MapInfo, 2006). The goal of
gaining more prominence with this segment has influenced Ekorne’s market-
ing, advertising, and distribution of resources. Several other geodemographic
segmentation products are PRIZM from Claritas and ACORN from CACI.
Geodemographic segmentation can be applied in spatial analysis, including
as follows (Harris et al., 2004):
1. Map the most prevalent segment for geographic units such as block
groups.
2. Compare the proportions of segments from a small area (say a ZIP code)
to the proportions of segments for a larger area (e.g. metropolitan area).
Figure 10.6 Community Tapestry LifeMode Summary Groups by U.S. County.

Copyright © 2006 ESRI. All rights reserved. Used by permission
3. Map indices of geodemographic features. For instance, the ratio of

a geodemographic feature to the national average can be mapped by
county.
4. Correlate a non-segmentation attribute with the proportion in a cer-
tain segment group for geographic units. For instance, cell phone calls
by census track can be correlated with percent of respondents in the
Tapestry category “Upscale Avenues.”
5. Statistically analyze a sample of zones that have common geodemo-
graphic features.
Geodemographic analysis is a powerful method based on enormous

databases and extensive pre-processing to determine the segmentation. How-
ever, it needs to be used cautiously, since it’s only as accurate and timely as
the underlying data used to create it.
A further weakness is that a segmentation is a generalization of a
neighborhood population which has considerable variation within it (Harris
et al., 2004). There also is the problem in geodemographics of the “ecolog-
ical fallacy.” This refers to analyzing a problem at one geographical level,
say census tract, and applying the results and explanations to another ge-
ographical level, say county. Geodemographic segments are initially deter-
mined for small areas, commonly block groups in the U.S. If the results
are applied for larger areas, the ecological fallacy cautions against interpret-
ing and explaining results for them. In spite of these critiques and weak-
nesses, geodemographics constitutes a large industry subsector and is used
extensively by mid- to large-sized companies for target and other marketing

decisions.
While target marketing provides improved product adoption for a product
or service at a certain time and place, adoption and diffusion theory tracks
the adoption and diffusion of a product or service over both time and space.
Originally developed by Rogers (1995), this theory postulates that initial
adoptions will be slow, then gain momentum quickly as the product diffuses
and more and more potential adopters learn and adopt it. Eventually the
diffusion will be over and the rate of new adoptions will approach zero.
Figure 10.7 plots new adoptions based on an adoption/diffusion process
divided into three stages, Stage 1 (lift-off), Stage 2 (rapid increase), and Stage
3 (leveling off). The boxes give the cumulative percentage of adoptions. This
process also takes place over space, starting with the location(s) where the
product is introduced, and spreading out. Consider a new car model that
appears simultaneously at 500 dealerships nationwide. Its adoption/diffusion
process can be compared to the Rogers theory (Figure 10.7) and its spatial
process studied. Initial buyers are located nearer to the dealerships. As more
and more potential buyers hear about the model, more distant buyers will
come to the dealerships and some will purchase. Eventually the market will
become saturated and the process will level off. This illustrates the process
of spatial adoption/diffusion.
To research spatial adoption/diffusion, extensive data must be gathered on
the adopters, the elapsed time from product introduction to adoption, and the
residential locations for adopters. Data need to be collected at regular inter-
vals, best collected daily or weekly for maximum insight. For example, daily
or weekly data is necessary if adoption diffusion takes place over a one-year
period. Data for the entire adoption/diffusion process can be analyzed by
geocoding the location of the central store and of all potential customers
(adopters and non-adopters) in the surrounding areas. Other topographic
features can be added as needed such as freeways or rivers, highways, bill-
board locations, and competing stores.
Stage One Stage Two Stage Three
Figure 10.7 Stages in Adoption-Diffusion Theory, Measured by Cumulative Per-

centage Adoptions. Source: This article was published in Journal of Retailing, Vol
79, A.W.Allaway, David Berkowitz, and Giles D’Souza, Spatial Diffusion of a New
Loyalty Program through a Retail Market, Pages 137–151, Copyright Elsevier 2003
The information is not easy to collect, as customers may not readily volun-
teer it, much less over a one year period. However, a firm is able to overcome
this by offering premiums or bonuses to customers in exchange for gathering
the data.
An example of this is adoption of loyalty cards by a very large retailer
in a major U.S. city, which was studied by Allaway et al. (2003, 2005).
The retailer made available to the academic researchers a database on nearly
18,000 adopters of loyalty cards, spanning a one-year period following the
loyalty card introduction. The researchers expanded the database with further
attributes that they calculated, including distances, locations of billboards,
and the extent of presence of early adopters in surrounding areas to see their
influence on spatial patterns and velocity of adoption (Allaway et al., 2003).
Very small geographic units were used, even smaller than the census block
group, namely the Neighborhood Information Field (NIF). This is defined as
“the geographic area within which a person has a relatively high probability
of contacts on a regular basis with other persons living around him or her.”
The area has a radius of 0.06 mile, and involves only thirteen to seventy-five
households (Allaway et al., 2003).
The three adoption stages assumed for the project are the following. Stage
1 is early adoption by a random group of consumers mostly near the center.
Stage 2 reflects the spread of adoptions away from the center and the presence
of the “neighborhood effect,” i.e. groups of new adopters that surround early
adopters (Allaway et al., 2003). Stage 3 is called the “saturation stage,” and
involves the filling in of gaps in the spatial pattern until the diffusion is finished
(Allaway et al., 2003). One month after introduction, the spatial diffusion is
summarized in Figure 10.8.
25.00%
Stage One:
Stage Two:
20.00%
Stage Three:
15.00%
Nonadopters:
10.00%
5.00%
0.00%
0-1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Miles from the Store
Figure 10.8 Retail Study of Loyalty Card Program—Percentage Distribution of
Adopter Groups by Distance from the Central Store. Source: This article was pub-
lished in Journal of Retailing, Vol 79, A.W.Allaway, David Berkowitz, and Giles
D’Souza, Spatial Diffusion of a New Loyalty Program through a Retail Market,
Pages 137–151, Copyright Elsevier 2003
Store
5.5
miles
Adopters
Nonadopters
Innovators
Figure 10.9 Retail Study of Loyalty Card Program—Pattern of Adoption Activity

5.5 Miles from the Central Store. Source: This article was published in Journal of
Retailing, Vol 79, A.W.Allaway, David Berkowitz, and Giles D’Souza, Spatial Diffu-
sion of a New Loyalty Program through a Retail Market, Pages 137–151, Copyright
Elsevier 2003
It’s clear that there was considerable range in the distribution of all the
adopter groups, with successive stages distributed farther from the center.
The non-adopters were the most spread out from the center of all the adopter
groups, with 5 percent still present 12 miles away from the center.
Using maps based on this model, it was possible to study the effect of
billboards, positioned at 2.6, 2.8, and 2.9 miles from the center. A positive
“billboard effect” was present based on differences between adopters and
non-adopters in the vicinity of the billboards (Allaway et al., 2003, 2005).
Given the small size of the NIFs, it was possible to study the spatial patterns
of adoption within NIFs, and clusters of adoption were apparent around the
earliest adopters, called Innovators. “At every distance from the store, a much
higher percentage of Stage 2 and 3 adopters had Stage 1 adopters within their
NIFs than for non-adopters” (Allaway et al., 2003). Figure 10.9 is an example
of the spatial pattern of adoption activity 5.5 miles from the central store. It
is evident that the Innovators (Stage 1 adopters) tend to have a grouping of
Stage 2 and 3 adopters closely surrounding them.
In short, this retail example of spatial adoption/diffusion shows how cumu-
lative daily customer data on a new product can be modeled in space based on
stages of adoption/diffusion. The model provides insight into spatial diffusion
in different adoption stages, the influence of the earliest Innovators, and the
importance of modeling physical advertising, in the form of billboards. The
approach requires a lot of data and a sophisticated model, but pays off in the
deep insights into micro changes in adopter and advertising roles in spatial
diffusion.
BANKING AND FINANCE 311
The Marketing/Geodemographics industry sector has a powerful range of

tools available for spatial marketing analyses. The approach taken depends
on the availability of data and the constraints on the amount of time and
resources that can be spent in developing models. Another challenge is to
educate executives about how to interpret the sometimes complex outputs.
In the U.S. there is wide availability of demographic data at no cost for
geodemographics and other marketing techniques that incorporate GIS. The
U.S. Census collects large amounts of data through the decennial census held
every 10 years. Hundreds of demographic, social, and economic attributes
are available for individuals and households. The data are available for many
geographical units including counties, metropolitan areas, cities, tracts, and
block groups, i.e. small areas of about 1,500 people.
Updated socio-demographic data are now also available on an annual
basis from the Census’s American Community Survey, a large three-year,
rolling national survey, and those data in 2010 will be available down to the
census tract level. All these data and much more including specialized surveys
are available for free and largely web-accessible from the Census website
(www.census.gov). The website even provides web mapping for selected
characteristics. Few other nations have this quantity and ease of access, but
many other countries provide substantial demographic data, especially at
census times, although often with costs. In all, marketers in countries with
modern censuses and policies of open access can usually find the data they
need for the variety of GIS-based marketing applications discussed.
BANKING AND FINANCE

Site selection and market planning
Facility property management
Branch and ATM network planning
Regulatory compliance
Geography matters for the Banking and Finance Industry in tracking and
locating customers, facilities, and financial assets. Even in the era of the
web and mobile technologies, physical sites remain crucial in developing
the personal relationships that have characterized success in this industry. The
case study example of the Large Personal/Corporate Bank (LPCB) in Chapter
3 illustrated how a rapidly expanding bank in the Northeast of the U.S. utilized
GIS for decision support in planning new branch locations. Banking and
finance also have to meet regulatory requirements with locational provisions,
for example regulation of lending to poorer and minority customers.
In site selection, GIS can model the often complex branch account struc-
tures and the distribution of branch services (ESRI, 2006). GIS can identify
areas that are underperforming in sales, or missing potentially high-value

customers. Spatial models can include transport times and distances, ob-
stacles such as rivers or mountains, competitor locations, and proximity
to urban amenities that attract customers. Markets surrounding branch or
ATM sites can be assessed and modeled using trade-area techniques. The
relationship of multi-channeling and the branch network can be spatially
studied.
An example is the Altura Credit Union, a small but fast-growing credit
union in Riverside, California, with 106,000 members and $750 million in
assets. Credit unions are open to certain allowable members, in this case
to a person who lives, works, attends school, or worships in Riverside and
San Diego Counties and parts of two others (Khan and Lackow, ). The bank
partnered with an outside banking consultant to build a GIS to analyze the
potential of prospective sites. In a branch study in 2005, the spatial analysis
pointed to three prospective sites for new branches, two in Temecula and
one in Murietta Hot Springs in southwestern Riverside County (Khan and
Lackow, 2006). An initial mapping of existing sites was achieved by mapping
a combination of population and geodemographic data available in ESRI’s
Business Analyst Online. Competitor information was then added using com-
mercially available branch information on competitors’ branches (Khan and
Lackow, ).
Next, maps and reports were produced on the three prospective sites. For
instance, the map seen in Figure 10.10 shows estimated market potential
levels for the three potential branches with thematic shadings ranging from
white (very low) to black (very high). It is overlaid with the competitor
locations shown as circles, with radii proportional to the amount of deposits.
Squares showing competing credit union locations are all of the same size,
since deposit information was not available. This map was one of several that
Altura used to narrow the potential new sites to three: Murieta Hot Springs,
Temecula/79, and Temecula/Ynez.
Although the GIS analysis indicated one of the Temecula sites was the
most favorable based on marketing-potential level, Altura ended up deciding
to locate the new branch at the Murieta location due to short-term real estate
considerations. However, at the same time, management planned to open
the GIS-favored Temecula/79 site a year later (Khan and Lackow, ). This
case of a small credit union demonstrates how GIS can assist in site decision
making through market and competitor analysis, even though in this particular
instance the final choice was based on managers’ gut experience and the
existing real estate opportunities on the ground.
CASE STUDY OF LARGE CREDIT BANK
Large Credit Bank (LCB) is one of the largest credit card banks in the U.S.,
with over $50 billion in deposits. It has not had branches until recently when
CASE STUDY OF LARGE CREDIT BANK 313
Figure 10.10 Altura Credit Union—Market Potential and Competitor Analysis for
Three Prospective Branch Sites in Temecula and Murrieta Hot Springs, California.
Source: Khan and Lachow,
it acquired two consumer banks in the South that utilize GIS for their branch
networks. LCB has depended for its IT on legacy mainframe systems. LCB’s
GIS is outsourced to a small outside GIS consulting firm. The bank’s primary
use of GIS is to map internal data for existing bank sites including customer
and competitor locations. Although demographic data are available, they are
not included in the siting maps. Locational mapping falls very short of its
potential to integrate internal and geodemographic data to study existing and
future sites relative to trade areas, competition, topography, and transporta-
tion.
The firm’s IT group is not accepting of GIS, making it even more difficult to
implement. A glimmer of hope comes from the newly acquired banks, which
already have GIS and are pushing for more of the technology. Due to resistance
to change, GIS is not regarded as strategic, but rather is relegated to minor
and simple locational studies. This case highlights organizational resistance
to technological change. It also illustrates how acquisition by a larger bank
of more technologically advanced consumer banks can potentially stimulate
openness to contemporary spatial approaches.
AGRICULTURE

Precision agriculture
Planning and management of crops on farms
Monitoring of livestock
Epidemiological analysis
Spatial systems are utilized in Agriculture for management, planning, pre-

cision treatment of fields, livestock monitoring, and epidemiological analysis.
They improve agricultural efficiency and productivity, allow better controls,
and track disease. The technologies impact stakeholders, including farmers,
agribusiness managers, government agencies, and regulatory authorities, and
investors.
Precision agriculture is the use of agricultural methods, treatments, and
information to farm micro sections of fields. The concept is to provide spe-
cialized treatments to areas in the range of several meters square. This is pos-
sible by sampling information for the tiny areas through sensors, yield moni-
tors, and remote sensing, utilizing large-scale digitized federal soil maps and
recording the precise locations through GPS. The model for precision agricul-
ture seen in Figure 10.11 starts with the input data from these sources, which
are stored in a GIS. The GIS processes them to plan out the management
zones. In other words, the micro-plot information allows the farmer to divide
the field into smaller sections based on differences in soil, plantings, and
micro-climate in order to apply the optimal micro-treatments. The GIS also
provides decision support regarding daily field treatments, and longer-term
field management. GIS can help optimize the farmer’s daily tractor routes that
provide the micro-treatments and can model monthly or yearly yield trends
for micro sections of the field. Mobile devices such as the ruggedized, in-cab
computer for a tractor or sprayer seen in Figure 10.12 provides the graphical
display, mapping, application coverage logging, and record keeping for an
automated steering system used for precision agriculture.
More specifically, field-oriented steps and tools that can be applied for
spatially-enabled farming include:
r Equipment guidance. A farmer can utilize NavStar-like equipment to

support accurate routing in fields. This equipment is referenced by GPS
and/or an inertial guidance system (Ping et al., 2004), and has guidance
software that allows it to minimize fuel, labor, and maintenance costs
(Adrian et al., 2005). For instance, in treating a field, the equipment can
reduce criss-crossing the same area or avert not covering a zone. Further,
it can lower environmental damages from pesticides and herbicides.
r Auto-steer system. It is a combination of a computer in the vehicle with
GIS software, a device that gives direction indication to the driver known
AGRICULTURE 315
Figure 10.11 Model of GIS in Precision Agriculture. Source: Adrian et al. (2005).
Copyright 2005, IGI Global, www.igi-pub.com, reprinted by permission of the pub-
lisher
Figure 10.12 Agricultural Computer for Field Operations and Decision-Making

(Trimble Ruggedized AgGPS 170). Source: Trimble, 2006
as a lightbar, and ground-referenced GPS that allows centimeter accuracy.

The auto-steering gives guidance accuracy to within two inches of a
location for planting and irrigation, lowering cost.
r Yield monitor. A yield monitor on a harvester or crop picker indicates the
volume of crops harvested, keyed to GPS readings and to sensor readings
on elevation and climate (Adrian, 2005).
r Soil sampling. The soil conditions on a micro basis profoundly influence
agricultural productivity. Soil samples can be taken meter by meter.
Sometimes soil sampling divides the field by square grids or by zones
based on federal soil maps (Adrian et al., 2005). Based on soil and other
factors, precision agriculture is able to apply water and fertilizer with
meter accuracy (Ping et al., 2004).
r Remote sensing. Remote sensing by satellite or airborne sensors pro-
vides information on soil moisture content, organic matter, and an es-
timate of plant health (Adrian et al., 2005). Although the method is
expensive, can be deterred by clouds, and requires careful overlaying at
meter accuracy, it can cover large areas and is amenable to longitudinal
comparison (Adrian et al., 2005). The remote-sensed images are some-
times useful to recognize areas with plant disease and insect damage
(Ping, 2004). As remote sensing is a relatively new approach, training is
essential.
The planning and management of fields and crops are helped by consolida-
tion of the rich database collected. Using decision support systems and expert
systems, a longer-term production plan can be developed. For example, a year
of production data at micro-level accuracy can be used to plan the next year’s
production (Ping, 2004).
An example is Cargill’s Palm Oil operations in Hawaii, formerly Pacific
Rim Palm Oil Ltd. (MapInfo, 2005; Cargill, 2006). The unit uses an integrated
agronomy management information system, to analyze agricultural data on
the plantations over time and space, covering “yield, leaf analysis, soil anal-
ysis, environment, climate, pest and disease, and palm census” (MapInfo,
2005). Map displays show the results of models that track crop production
and other performance indicators. Satellite images are overlaid on attribute
maps to provide highly accurate physical features. The managers consider
the spatial application user-friendly. The benefits are higher palm oil produc-
tion, better management of crops, and mitigation of environmental damages
(MapInfo, 2005).
A different kind of spatial application involves livestock. Diseases in live-
stock can be monitored through use of RFID equipment (Stewart, 2005). This
is important in controlling the spread of livestock-carried illnesses. Obstacles
include inaccurate reading of RFID sensors attached to very large livestock
such as cattle, and fragility of RFID readers in the context of cattle and
HEALTH CARE 317
other livestock handling. The coordination of disease outbreaks, done jointly

between private companies and the government, is advanced by the detailed
inventory and GIS tracking that RFID makes possible.
In sum, spatial technologies are useful in agriculture as they monitor and
treat agricultural events on the ground, assist in managing and planning farm
production, and track and mitigate disease.
HEALTH CARE

Siting, access, catchments
Facilities management, bed control
Epidemiology and disease management
Health emergency services
Marketing
Forecasting
The Health Care industry has many locational applications. Health Care is
a large economic sector that comprises a mixture of intertwined public, pri-
vate, and nonprofit organizations. The industry can be viewed as a value chain
that starts with manufacturing, including pharmaceutical, hospital supply, and
hospital construction; progresses to intermediaries such as equipment distrib-
utors and pharmaceutical warehouses; and ends up with service providers
such as hospital chains, pharmacy chains, managed care facilities, and elder
care communities. Many general services are provided to the value chain, such
as market consultants and lending institutions. Baystate Health, covered in
Chapter 2, is a for-profit health-care network that applies GIS to emergency
planning, epidemiology, routing, marketing, facilities planning, emergency
services, and surgery. Baystate’s GIS champion, a surgeon, gave emphasis
to its medical and health benefits. Other health-care enterprises have spatial
applications more weighted towards the operational and business aspects, as
in the Kaiser Permanente case.
A common spatial application in health care is to determine siting, access,
and catchment areas. GIS analyzes the siting of hospitals, clinics, physicians,
patients, equipment, and members of networks. This can be performed based
on methods to identify trade areas, including the Huff model covered in
Chapter 9. Analysis can show the best locations with respect to accessibility,
demographics, and avoidance of cannibalization (Thrall, 2000).
For instance, Jewish Hospital and St. Mary’s HealthCare Services
(JHSMHS) is the service wing of a regional health-care network of 70 health
care facilities and 1,900 patient beds in Kentucky and Southern Indiana
Jackson
Switzerland
Jefferson Pendle-
Scott Grant ton
33 Carroll
33 Trim-
Orange Washington Clark blg
Owen Harrison
Henry
Oldham
Floyd 333111 Scott
Crawlord 3433 Bourb
25 B Franklin
Harrison Jefferson
Perry Wood-
2 17 Spencer fore Fayette
11 4 W Anderson C
Bullitt Jessa-
Mercer
mine
Nelson Madiso
Breckingbridge Washington
Hardin Boyle Garrard
Marion
Larue
Lincole
Grayson 34 31 Rockcastl
Figure 10.13 Jewish Hospital and St. Mary’s HealthCare Locations. The map
shows county names and an inset for Jefferson County that includes Louisville,
Kentucky. Source: JHSMHS Website, 2006
(JHSMHS, 2006). JHSMHS serves the health-care locations with 24-hour

treatment services for specialized rehabilitation and preventative health-care
programs such as employee alcohol and drug screenings. The spatial chal-
lenge for JHSMHS was to analyze where patients and employers were located
relative to JHSMHS facilities. Maps were developed to show the health
centers, service catchment areas, and patient locations. Figure 10.13 shows
JHSMHS facilities including the major hospitals at Louisville along the Ohio
River in northwest Jefferson County and others in a fifty mile circle around
Jefferson County encompassing 40 other Kentucky and Indiana counties. The
numbered symbols represent hospitals, medical centers, rehab facilities, and
clinics, while the unnumbered ones are home care facilities and service areas.
Location analysts can use the GIS system to recommend new locations that
have a sufficient patient base and yet do not cannibalize the patient base of
existing sites (Lang, 2006). Examining the number and location of employers
in that service area is also useful for understanding the typical type of
injuries.
GIS helps in managing facilities. The layout of hospitals and medical
centers can be visualized and modeled. In hospital bed tracking, since beds
are often difficult to locate and also expensive, costs can be reduced through
a GIS that can model the bed occupancy over time to identify more efficient
bed locations and groupings for certain patients. For example, MapGistics is
a small company that provides bed management consulting. Its “Enterprise
Bed Management System” product mimics the whiteboards that nurses use on
floor stations to designate patient locations. Using GIS it is able to provide
maps of hospital floor plans showing bed occupancies A benefit of more
SPATIAL TECHNOLOGIES AND OUTSOURCING 319
accurate bed management is that discharge times are more accurately known,
and their sub-steps can be better planned. As a result, discharge takes about
half the time it did prior to automating bed management.
In epidemiology, maps are utilized to track and monitor disease spread,
and better understand the causes of diseases. Point maps show the locations
of patients with particular illnesses. The amount of clustering or separation
of ill patients helps to diagnose cause and determine where treatments are
working. Since certain illnesses are more likely in susceptible environments
GIS can map those locations (Thrall, 2000). For certain infectious diseases,
crowded and high-density urban places stimulate communicability. Maps can
indicate areas of high risk (Thrall, 2000).
When health emergencies occur involving transportation, spatial technolo-
gies are significant in real-time tracking, communications, and planning. The
reader is referred to the Transportation section of Chapter 9, on spatial tools in
emergency transportation. In emergency health care transport speed and ac-
curacy of routing are essential and even a matter of life and death. Optimized
route management software complement real-time databases indicating the
patient’s condition and the availability of medical services at different desti-
nation locations.
Marketing permeates many health-care organizations. Spatially-enabled
marketing tools described earlier in the chapter under Marketing/
Geodemographics can provide more effective marketing to targeted markets.
For instance, a managed care company can market its membership benefits by
targeting its advertising to small areas dominated by certain geodemographic
segments.
An example of GIS-enabled marketing is Health Products Research Inc.,
a division of Ventiv Health Inc. (Lang, 2000). The firm offers a model that
predicts how many physician visits are necessary by a field representative,
in order to achieve cost-efficiency. It is based on data on all prescriptions
written by U.S. physicians in the last two years. This model of sales potential
for physician prescription writing on a national basis is used by drug company
sales reps. The maps guide sales campaigns and show geographical areas in
the U.S. with deficits in detailed prescription-writing and opportunities on a
cost-benefit basis (Lang, 2000).
SPATIAL TECHNOLOGIES AND OUTSOURCING
Outsourcing of information technology is a major factor in business today

(Willcocks and Lacity, 1998; Lacity and Willcocks, 2001). There has been
enormous growth in IT outsourcing worldwide. The total size of IT outsourc-
ing in 2007 is estimated by Gartner Inc. at $236 billion with a compound
growth rate from 2002 to 2007 of 7.8 percent (Caldwell et al., 2004). By dol-
lar value, the segments of outsourcing in 2007 are data center (39 percent),
network (28 percent), enterprise applications (21 percent), and desktop (12
percent) (Caldwell et al., 2004). GIS outsourcing is a growing part of IT
outsourcing (Daratech, 2005). Its growth is changing the way companies or-
ganize, staff, and conduct spatial projects. This section looks at key factors in
IT outsourcing that also apply to GIS outsourcing, and presents a short case
study of outsourcing for a U.S. utility firm to an offshore outsourcing provider.
Outsourcing is defined for GIS as the provision to an outside party of
the management of GIS activities, resources, and assets, in order to achieve
a result required by the requesting firm. A similar definition applies for IT
outsourcing by substituting “IT” for “GIS” (Willcocks and Lacity, 1998).
Outsourcing involves the contracting of the GIS responsibility, either on a
project-by-project basis or in a long-term relationship as a preferred con-
tractor. Outsourcing is different than hiring temporary personnel or having
a regular firm that provides contract programmers on call. Contracting firms
provide contract employees that work in-house and report to in-house man-
agers, but the contracting firms do not take outside responsibility for managing
projects and functions (Willcocks and Lacity, 1998). Instead the firm that hires
the contract employees maintains the management responsibility.
Why do organizations undertake outsourcing? One study indicates four
underlying reasons: (1) cost reduction and containment, (2) organizational
restructuring, (3) management desire to rid the firm of direct management
of a “troublesome” and unprofitable part of the business, and (4) conformity
to industry trends i.e. do what others are doing (Willcocks and Lacity, 1998;
Earl and Feeny, 1994).
IT outsourcing varies along two dimensions: local versus global, total ver-
sus selective. Local outsourcing is a service in the local area. Since there is
potential for closer physical interactions, it often suits the needs of smaller
firms. Global outsourcing applies to larger firms. One study of large orga-
nizations indicated that the proportion of outsourcing transactions for local,
regional, and global was 60, 15, and 25 percent respectively, while by average
outsourcing transaction size was biggest for regional ($378 million), followed
by global ($210 million) and local ($136 million) (Everest Research Institute,
2004). Thus the proportion of transactions locally would be much higher for
small firms that outsource some of their work
A firm can contract their outsourcing totally to a provider, in which case
they only retain management oversight. Alternatively a firm can contract
selectively with the provider, retaining some aspects internally. Selective out-
sourcing has been shown to be the most common approach for IT (Lacity,
Willcocks, and Feeny, 1995). Another study found that short-term selec-
tive outsourcing is a very successful approach, and is preferred to long-term
arrangements (Lacity, Willcocks, and Feeny, 1995). Multi-sourcing, i.e. se-
lective outsourcing to multiple providers that work together (Cross, 1995),
has become somewhat common for large companies. It involves a trade-off
between the risk of escalating prices relying on a single provider versus the
expanded coordination cost of utilizing multiple providers (Cross, 1995).
KAISER PERMANENTE 321
Multi-sourcing sometimes involves outsourcing IT to one vendor and GIS to

another. In sum, IT outsourcing is large and growing in business. It can be done
through varied arrangements, time periods, and proximities of the provider.
A key issue is the cost and sometimes the bother of having IT in-house versus
the coordination costs and loss of control of having it outside.
In GIS, there are many outsourcing services, of varied types. Some offshore
providers in GIS offer the lowest cost. However, given the complexity of GIS,
client companies often will not opt for the lowest cost but seek the data
quality, accuracy, experience, and capabilities to handle complex projects
(Daratech, 2005). Utilities have tended to outsource non-core GIS applications
(GeoWorld, 2004). The arrangements vary from total to selective outsourcing
of non-core GIS, with most core GIS work done in-house. Some utilities
allow the outsourcer to perform some of the IT processing and upload the
results into special web portals for the utility’s asset management and work
management systems (GeoWorld, 2004).
As discussed in Chapter 6, LCB and Kaisen among the book’s cases
outsourced GIS significantly but not completely, while Sears & GIO had
limited selective outsourcing. The case experiences with it were generally
positive.
Another example of GIS outsourcing is Southern California Edison (SCE).
The provider Patni is based in India and has over 15,000 employees (Dighe,
2005). Patni has 300 people on its GIS team in India, who support the leading
GIS software packages. These employees receive GIS training from ESRI
India or in-house. SCE uses ESRI software and a utility-specific commercial
package.
Although usually Patni has allocated far more of its GIS people in India
than in the U.S., the SCE contract specified that there would be a dozen Patni
people assigned to SCE facilities in the U.S. and only five in India. The Patni
people at SCE have the identical GIS software, interfaces, and data as the
Patni staff in India, so groups can communicate better during projects. Also,
Patni in California can work in the middle of the day, while Patni in India is
asleep, so work occurs at all hours. Patni had to learn about some business
practices culturally different from India such as who participates and speaks
up at meetings and where authority rests. Lower-level projects such as data
conversion and GIS software maintenance can take advantage of a much larger
and less expensive pool of GIS talent in India, while SCE employees can focus
on planning and high-level design. In sum, GIS outsourcing at SCE is working.
Patni communicates well with SCE and responds to cultural differences.
KAISER PERMANENTE
Kaiser Permanente is a non-profit and one of the largest health-care organi-

zations in the U.S. Founded in 1945, it has approximately 12,000 physicians,
431 medical offices, and thirty medical centers. Its 8.4 million members who
enroll voluntarily are mostly located in California, but some are in eight other
states. It emphasizes cost-effective integrated health care.
GIS is utilized in travel-time analysis, site selection, long-term forecasts,
regulatory information, in-house reference maps, spatial websites to serve
members, support for an ambulance service, sales and marketing, and fi-
nance/accounting.
Most Kaiser members live in urban areas with potentially high levels of
traffic congestion. To gauge access to medical services, members need to
know the travel times to Kaiser offices and centers. The travel-time spatial
application computes rush-hour travel times for specific types of services in
urban areas. For instance, the estimated travel time to access end-stage renal
disease treatment sites are available for parts of San Francisco’s East Bay
(Figure 10.14). ESRD (End State Renal Disease) sites are shown, as well as
MOB (Medical Office Buildings) and hospitals. Although most of this area is
well covered, some hilly areas in the middle of this figure have 30+-minute
drive times, which can present problems for patients with kidney failure.
Figure 10.14 Kaiser Permanente Estimated Travel Times to End Stage Renal Dis-
ease Treatment Sites, 2005. Source: Schoenhaut, 2005
CHAPTER SUMMARY 323
This travel-time application was developed over many years through a

combination of in-house design and outside consultants. In other words,
some of its development was outsourced. For instance, GIS consultants in
Oakland, California, Hammond, Jenson, and Wallen, developed a minimum-
travel-time matrix. Another outside group from Cal Trans drove the routes
to determine travel times. Recently consultants from the Berkeley Institute
of Transportation Studies have focused on design of a network of detectors
to measure vehicle speeds. Based on the data purchased from TeleAtlas, the
travel-time applications give travel speeds for the morning and afternoon rush
hours. The approach was selective outsourcing, combining some tasks better
done in-house with others better outsourced.
Site selection is analysis of areas to arrive at the best locations for services.
The goal is to maximize the number of members with access. For instance,
in northern California Kaiser conducted a Radiation Therapy study using
location-allocation modeling that identified the fifteen best radiation therapy
sites from over forty potential sites (Schonhaut, 2005).
Fifteen-year forecasts are performed for medical services and membership
at the ZIP-code level. For instance, the map of Kaiser Permanente’s forecasted
member densities can be overlaid on the Kaiser facilities and current service
areas (Figure 10.15). These studies help middle managers and executives
strategize future growth.
Kaiser’s spatial web applications consist of an intranet-based library of
several hundred reference maps that are widely accessible by employees
across the Kaiser system. The GIS team has developed limited web services
for members, including access maps. Mapping is done of ambulance pick-ups
for all of the Kaiser emergency cases in San Francisco that are sent to non-
Kaiser hospitals. In San Francisco, regulatory restrictions on Kaiser force
members to seek services out-of-network. As some patients visit non-Kaiser
emergency rooms this system allows patients to be tracked and their records
be retained in the Kaiser database.
CHAPTER SUMMARY
This chapter continued the previous chapter’s focus on industry differences

and organizational aspects, presented five additional industries which are
classified as less mature in GIS than those covered in Chapter 9, and dis-
cussed GIS outsourcing. Each industry has critical specialty areas—for ex-
ample, climate/disaster management for insurance, segmentation for mar-
keting/demographics, branch and facilities locations for banking, precision
agriculture, and travel time analysis for health-care companies. Banking is
illustrated with the LCB Case Study, while health care services is exemplified
by the Kaiser Permanente case. While GIS outsourcing provides the benefits
Figure 10.15 Kaiser Permanente Forecasted Density of Membership in Service

Areas of East Bay, Greater Southern Alameda, and Part of Diablo, 2005. Source:
Schoenhaut, 2005
of lower cost and larger pools of specialized talent, there are weaknesses in
increased coordination costs, communication problems, and loss of control.
REFERENCES
Adrian, Anne Mims, Chris Dillard, and Paul Mask. 2005. “GIS in Agriculture.” In
Pick, James B. (Ed.), Geographic Information Systems in Business, pp. 324–342.
Allaway, Arthur W., David Berkowitz, and Giles D’Souza. 2003. “Spatial Diffusion
of a New Loyalty Program through a Retail Market.” Journal of Retailing, 79:
137–151.
Allaway, Arthur W., Lisa D. Murphy, and David K. Berkowitz. 2005. “The Geo-
graphical Edge: Spatial Analysis of Retail Loyalty Program Adoption.” In Pick,
REFERENCES 325
James B. (Ed.), Geographic Information Systems in Business, Hershey, PA: Idea

Group Publishing, pp. 260–277.
Anselin, Luc, Raymond JGM Florax, and Sergio J. Ray (Eds.). 2004. Advances in
Spatial Econometrics: Methodology, Tools, and Applications. New York: Springer.
Bedell, Craig. 2006. “Location Technology Directs Insurers To Greater Efficiency,
Higher Profitability.” National Underwriter Property and Casualty, January 20.
Bourgault, Steve. 2005. “Target Marketing and the Spatial Advantage.” MapInfo
Magazine, 9(1): 18–30.
Cross, J. 1995. “It Outsourcing: BP’s Competitive Approach.” Harvard Business
Review, 73(3): 94–104.
Earl, M., and D. Feeny. 1996. “Is Your CIO Adding Value?” Sloan Management
Review, 37(3): 26–32.
ESRI. 2006. “Insurance Overview.” Redlands, CA: ESRI Inc.
Foust, Brady. 2006. “Integrated Enterprise Geo-Spatial Technology—Insurance Risk
Examples.” Directions Magazine, January 27. Available at www.directionsmag
.com.
Harris, Richard, Peter Sleight, and Richard Webber. 2005. Geodemographics, GIS,
and Neighborhood Targeting. Hoboken, NJ: John Wiley and Sons.
Khan, Adam, and Steve Lackow. 2006. “Case Study: Developing Market Intelligence
in Very High Growth Areas.” Proceedings of ESRI User Conference, Redlands,
CA: ESRI Inc.
Lacity, Mary C., and Leslie P. Willcocks. 2001. Global Information Technology
Outsourcing: In Search of Business Advantage. Chicester, England: John Wiley
and Sons.
Lang, Laura. 2000. GIS for Health Organizations. Redlands, CA: ESRI Press.
MapInfo Magazine. 2006. “Alliance Story: Siebel Systems and MapInfo Combine to
Offer Insurance Solution.” MapInfo Magazine, pp. 9–12.
MapInfo Magazine. 2006. “Florida Farm Insurance Case Study.” Troy, NY: MapInfo
Corporation. Available at www.mapinfo.com, September 1.
MapInfo Magazine. 2006. “RVOS Insurance Case Study.” Troy, NY: MapInfo Cor-
poration. Available at www.mapinfo.com, September 1.
Morton, Kimberly. 2002. “Insurance: Leveraging Location is Core to Managing
Risk.” MapInfo Magazine, 7(2): 3–11.
Ping, Wang, Xiang-nan Liu, and Fang Huang. 2004. “Research on Mobile Mapping
System and Its Application in Precision Agriculture.” Map Asia 2004 Conference
Proceedings, Beijing, China: GISdevelopment.net.
Rogers, E.M. 1995. Diffusion of Innovations. 4th Edition. New York: Free Press.
Sleight, Peter. 2004. “Targeting Customers: How to Use Geodemographic and
Lifestyle Data in Your Business.” Henley-on-Thames, UK: World Advertising
Research.
Steward, May Ann. “ Tracking Cattle in the Heartland.” Geospatial Solutions, 15(9):
20–25.
Swinton, S.M., and J. Lowenberg-DeBoer. 1998. “Evaluating the Profitability of
Site-specific Farming.” Journal of Production Agriculture, 11(4): 439–446.
Thompson, Paul. 2005. “Closing the Loop Between Real Estate and Marketing.”
MapInfo Magazine, 9(1).
Thrall, Grant Ian. 2000. “The Future of GIS in Public Health Management and
Practice.” Geospatial Solutions, September, pp. 2–7.
Villalon, Maria. 1999. “GIS and the Internet: Tools that Add Value to Your Health
Plan.” Health Management Technology, 20(9): 16–8.
Weiss, M. 2000. The Clustered World. New York: Little Brown.
Willcocks, Leslie P., and Mary C. Lacity. 1998. Strategic Sourcing of Information
Systems. Chicester, England: John Wiley and Sons.
CHAPTER 11
ETHICAL, LEGAL, AND SECURITY

ISSUES OF SPATIAL TECHNOLOGIES
INTRODUCTION
GIS and spatial technologies have provided remarkable monetary and other
benefits to organizations in the case studies in this book. At the same time,
GIS has sometimes failed from a financial standpoint to the detriment of
businesses, investors, and customers. This chapter reaches beyond business
benefits and costs, and considers ethical and legal questions involving these
technologies. Consider for example geodemographics from Chapter 10. The
technique provides information profiles about small neighborhoods and areas.
In doing so, it averages the information into a common profile. There may
be people living in a neighborhood who are different from the profile of say
“Connoisseurs” or “The Elders,” to use categories from ESRI’s Community
Tapestry segments, and they may be at times misrepresented by being merged
into a profile. There is an ethical issue in how far geodemographics should
be pushed by companies in marketing, telemarketing, advertising, credit, and
financing. After a brief general background on GIS ethics, the first section
considers examples of ethical dilemmas. The goal is not to prescribe behavior
or business policy, but to raise awareness of ethical issues.
Privacy is the right to be left alone, in the absence of reasonable grounds for
interference. Such grounds might be illegal activity or instances where a nor-
mal person would not expect privacy. One of the best examples is one’s home,
which in some countries including the U.S. is constitutionally assured of

327
328 ETHICAL, LEGAL, AND SECURITY ISSUES OF SPATIAL TECHNOLOGIES
privacy rights. Since GIS and spatial technologies pinpoint locations, sense the
environment, and record movement of people and objects, they can challenge
individual privacy. In a broader context, major changes in spatial technologies
have changed the entire notion of privacy, eventually leading to altered legal
interpretations of it. The privacy section begins with a short background and
proceeds to spatial examples, where privacy is threatened.
The law has long recognized concepts of place and space, way prior to
the advent of commercial GIS in the 1960s. Today, traditional legal concepts
are challenged by GPS, RFID, sensors, satellite imagery, and GIS. This is
analogous to the changes in criminal law that resulted from widespread labo-
ratory capability to analyze DNA in legal evidence. After presenting a short
background on law, the section proceeds to examine some examples with par-
ticular emphasis on legal and ethical controversies involving geosurveillance,
i.e. satellite or airborne imagery, and how some of them have been resolved.
Law for GIS and spatial technologies is evolving, and many more legal issues
will appear as these technologies become more prevalent.
The chapter case study examines a well-known legal case of geosurveil-
lance, Kyllo v. United States 533 U.S. 27 (2001). It involved an illicit small
business where the legal basis of geosurveillance detection was challenged in
a series of court decisions.
The last part of the chapter considers security issues of spatial technologies.
How can organizations assess their GIS and spatial security problems and
set up means to reduce security threats? This chapter raises issues, many
of them important to businesses with spatial systems and products, but is
usually not able to definitively answer them. The goal of the chapter is to
encourage awareness of these issues among concerned people in businesses
and organizations, and the stakeholder community.
ETHICAL ISSUES
Ethical concepts and theories are useful as background to the practical exam-
ples to follow. Only three categories of theories are briefly mentioned here:
(1) ethical relativism, (2) utilitarianism, and (3) deontological theories. Other
ethical theories and theoretical background (Johnson, 2001) are beyond the
scope of this book.
Ethical relativism is based on the assumption that ethics is relative. In
one’s society, there are no universal rights and wrongs, nor are there uni-
versal moral norms. Thus two people may differ in their ethical views, and
both views can be considered valid. Although open to diversity of view-
points, ethnical relativism has a number of problems including that the theory
has limited logical arguments to justify it. Another problem with this the-
ory is that there cannot be prescriptive claims, i.e. “you ought to do this.”
Since cultures simulate relativistic differences it is hard to apply it across
cultures.
ETHICAL ISSUES 329
Relativism relates to GIS and spatial technologies by supporting differ-

ences of opinions, but leaving the differences unresolved. For the fast-moving
spatial technologies, relativism may be applicable, since settling on fixed eth-
ical rules may not be durable if the environments are turned upside down by
new innovations. On the other hand, the fixed rules might be interpreted or
applied differently in a new technological context.
Utilitarianism determines right and wrong depending on the consequences
(Johnson, 2001). Behavior and actions are good, if they imply that happiness
dominates in the consequences. This can apply to the choices that have been
often made, for example, between paper mapping and GIS. Utilitarianism
can justify choice of GIS over mapping, arguing that GIS leads to greater
happiness. Since spatial technologies often have many stakeholders (users,
managers, investors, vendors, customers), it may be a complex challenge,
in a spatial decision, to determine what leads to greater happiness and for
whom. Utilitarianism results in more consistent rules than relativism, and
holds together better across cultures.
Deontological theories emphasize the inherent character of an act. A person
who adheres to these theories has a sense of “duty.” By contrast to relativism
and utilitarianism, these theories support that some actions are always right or
always wrong, even if the cost/benefit ratio is not favorable to an individual,
business, organization, or society. An important tenet of deontological theories
is that humans have the rational capacity to think and decide, in weighing
ethical choices. Although there may be decision-making that identifies a
phenomenon as a means to an end, that’s not the sole basis of the decision,
but other rational thinking must go into it.
Deontologists would indicate that happiness is not the highest good for
people. Rather the highest good relates to rules, moral principles, and laws
(Johnson, 2001). Deontological theories give value and respect to other hu-
mans, who are trying to rationally decide on ethical problems.
This class of theories also applies to ethical issues of spatial technologies.
An example would be for a delivery firm to not just base the decision to
adopt GPS devices for its entire fleet on the means to the ends of greater
efficiency and more profits, but to consider the rational moral principles. Is it
morally right for a driver’s location every minute during his work career to
be recorded? GIS codes of ethics that have been brought forward in the GIS
profession based on deontological theories are discussed later.
The final ethical concept to introduce is that of dialectic. An ethical dialectic
is the process of moving back and forth between theory and practice. The
process is a slow one, that may involve slight changes, but cumulatively
it leads to better ethical justifications. An example from the GIS field is the
ethical issue of the piracy of GIS software. The initial concept (deontological)
might say it’s never morally right. However, what about in an earthquake or
disaster, when more copies may need to be made very quickly to save lives.
The dialectic would allow the change in practice of copying to move back to
theory for possible modification.
A wide range of ethical dilemmas and issues regarding GIS and spatial
technology apply to businesses and their stakeholders. In this section some
will be mentioned, followed by deeper discussion of RFID devices. Among
the issues are the following:
(1) The ethical question of what map accuracy is best for the customer.
This has been a controversy between surveyors and GIS professionals.
Some sub-industries and professions demand high positional accuracy,
including commercial real estate, architecture, land development, and
civil engineering. Both surveyors and GIS analysts have certain compe-
tencies in map accuracy, but they compose and produce maps through
quite different procedures (Butler, 2005). Map accuracy in fact involves
three concepts: (a) accuracy, i.e. how close a mapped observation is
to true position, (b) precision, i.e. can the locational measurement be
repeated, and (c) resolution, i.e. how close together do two observations
need to be before they merge as one (Butler, 2001). Both surveyors
and GIS professional strive for all three, but they use different meth-
ods. There is the ethical question of whether professional “territory” is
more important or providing the highest type of accuracy to solve the
customer’s problem.
The ethical issue is whether to fully disclose these different profes-
sional claims and the full aspects of accuracy to clients or customers.
The issue is further complicated by legal requirements in certain states to
have one or the other type of professional assigned for certain mapping
tasks (Butler, 2001).
(2) Internet map services such as Google Earth and Yahoo Maps show maps
and satellite imagery for a large portion of the globe at a resolution that
reveals building structures and even large vehicles. It allows scrutiny of
land surface areas that some governments are averse to having as public
knowledge, as well as other types of secure areas. This leads to an ethical
issue of full freedom of information and service to customers versus
respecting the security concerns of certain countries. This issue may
appear settled on the side of the service providers and customers, but
will become a larger issue for the private citizen if the resolution of map
imagery improves even more for the public. It may become a domestic
issue for the private citizen if more detailed features of homes and other
private places become public knowledge. Such imaging may challenge
the “reasonable expectation" of privacy that an individual has, say in his
or her fenced backyard.
(3) Geodemographics characterize neighborhoods and small areas, so ev-
eryone in a small area might be consolidated together to have one profile.
This becomes an ethical issue if the representation of the individual af-
fects eligibility for economic and social benefits such as credit.
ETHICAL ISSUES 331
(4) Issues in GIS map design, in particular visualization that exaggerates

the dimension of particular features to make them more prominent. An
example is an environmental map from an activist group that emphasized
the noise impact from freeways by increasing ten-fold the width of
freeways on the map relative to other features. Although it makes the
point better for clients, it raises ethical questions of responsible and fair
display of information.
These four examples suffice as an introduction to the variety of spatial

ethical issues. The issues surrounding RFID are considered more deeply.
The uses of RFID are raising many contemporary ethical dilemmas. As
seen earlier in the book, RFID tags can be affixed to business high-value items,
inventory items, vehicles, animals, and people. Some common applications
are U.S. military inventory, business inventories such as Wal-Mart, vehicles
moving past toll booths on freeways, taking school attendance, monitoring
medical patients, checking and counting livestock, reading drivers’ licenses
in some states, and since 2006 reading U.S. passports. Ethical issues occur
when businesses consider when to dispose of the RFID tags in the supply
chain. A decision to leave tags on for-sale or sold merchandise raises ethical
issues of full disclosure and potential for privacy invasion. For instance,
a retail merchant might leave tags on durable items sold, without telling
the customer, in order to have information on the customer and transaction
available to the merchant (or anyone else with a portable RFID reader) in the
future.
Regarding RFID tags for people, there are even more worrisome emerging
ethical issues. People can carry or wear RFID devices, but an RFID tag
also can be injected by a syringe under a person’s skin (Havenstein, 2005),
where it is durable for up to 100 years (see Figure 11.1). The procedure
cost only about $200 in 2005 (Bradley, 2005). As discussed in Chapters 1
and 5, the chip in an RFID tag contains a lengthy and unique ID number.
That number is the key to accessing a computer-based record that can be
of unlimited length. For inventory, the ID number commonly identifies the
pallet and lot, but a longer ID also identifies the individual item. For a person,
the ID tag is a key that uniquely identifies a computerized record for the
individual.
In health care, the implanted RFID chip can be scanned at locations in
a hospital or other health facility, providing access to the patient’s medical
record. For example, the patient’s record can be scanned on check-in to a
waiting room, increasing efficiency. If the patient receives a test in a hospital
area, his/her full record can be available on arrival of the patient in order
to administer the test under correct protocols. At the same time, at each
RFID read point, the patient’s spatial location is recorded. Consolidating the
information from multiple patients, the mapping and spatial analysis of the
patient population and its movements in the hospital can be performed.
Figure 11.1 Close-Up View of VeriChip RFID Tag for Injection under Patient’s
Skin. Source: Graafstra, Amal; RFID Toys: Cool Projects for Home, Office, and
Entertainment; John Wiley & Sons, 2006, reprinted by permission
The FDA approved the uses of the implanted RFID tags for patients in Oc-
tober of 2004. There are several constraints to their use. One is the patient’s
fear of being injected with an unfamiliar device that appears to intrude on
privacy. A second problem is the paucity of RFID readers at hospitals and clin-
ics. Presently, a person injected with an RFID chip has spotty availability of
RFID readers. John Halamka (see Figure 11.2), Chief Information Officer of
both Harvard Medical School and Beth Israel Deaconess Medical Center and
emergency room physician, is an example of an early adopter, having received
his VeriChip implant in late 2004 (see Figure 11.3). Dr. Halamka sought to
get familiar with the device for his roles in medical IT planning and his own
patients. He also has a personal reason, since he is an avid mountain climber,
so the device might be life-saving, assuming that the doctors who might care
for him have access to an RFID reader and can link through the internet to
his record at Beth Israel Deaconess Medical Center (Havenstein, 2005).
As the world’s population ages, the elderly will become more preva-
lent, many of whom are debilitated and have problems with simple daily
tasks. RFID tags can be placed at key locations in the living areas of
the elderly, and the elderly person can be outfitted with a small, portable
RFID reader on a glove or necklace. As seen in Figure 11.4, a home or
health-care facility can have tags placed on key items such as a toothbrush,
doorway, toilet seat, and tea cup. As the person moves around in his/her
daily activities, the wearable RFID reader activates tags in the environment,
showing the daily individual’s daily movements and indicates activities, as
well as any glitches or problems that occur. For hospital patients who have an
implanted tag, RFID readers record their whereabouts, so patients throughout
ETHICAL ISSUES 333
Figure 11.2 John Halamka M.D., CIO of Harvard Medical School. He was im-
planted in 2004 with a VeriChip RFID Chip. Source: John D. Halamka
Figure 11.3 John Halamka M.D. of Harvard Medical School Viewing His RFID
Implant in 2004. Source: John Halamka, 2007
RFID tags are located at key ambulatory points in the building and room environment.
Individuals are outfitted with portable RFID readers
RoamAlert System includes:
RFID Asset
Visualization of Device to Tag to
locations of tags send distress locate and Real time
and alarms alerts and protect tracking of
gain access valuable infants, elderly,
through medical and patients
protected items, such
doorways as heart
monitors
Figure 11.4 VeriChip Roam Alert System. Source: Modified from VeriChip Cor-
poration, 2007
a hospital can be located. For patients and the elderly, RFID offers quick and
effortless recording of personal information and location.
However, ethical issues have been raised by implantation of tags in patients
and use by the elderly. Implantation is opposed by groups both on the basis
of ethical issues of privacy and religious concerns. For instance, the medical
history of the patient is automatically known to any RFID reader able to
read it, which would be restricted to hospitals and medical facilities, but
is exposed to being read by unauthorized persons. The spatial locations of
patients could potentially be tracked, without the patient’s knowledge. For
the elderly, their movements and activities are tracked and stored, which
some say compromises their privacy. One way that these ethical issues for
patients or the elderly could be mitigated would be regulatory or policy
changes requiring informed consent. There are some religious groups that
are adamantly opposed to implantation of RFID devices. Several Christian
groups claim it conflicts with the book of Revelation, which they indicate
states that Satan will force people to have a mark placed on their foreheads
or hands in order to engage in buying or selling.
ETHICAL ISSUES 335
Another RFID example is the Speedpass tag at Exxon-Mobil, which was

made available starting in 2004 to 10,000 ExxonMobil gas stations and over
seven million customers. Speedpass is an electronic-security-key device with
an RFID ID that links to the customer’s credit-card record. At the gas pump,
the customer places the RFID tag in front of a special reader, which automat-
ically deducts charges from the user’s credit-card account (Scheeres, 2003).
Although the tag’s ID was encrypted, the encryption algorithm is vulnerable.
It was broken by a Johns Hopkins professor and students from its Informa-
tion Security Institute. As an exercise, they were able to copy the Speedpass
tag and use it to buy gas. Speedpass has responded by taking increased
security precautions including asking customers for ZIP codes at time of
purchase.
Like Speedpass, Paypass is an RFID tag, in this case embedded in cer-
tain MasterCards. It allows a customer to make purchases by wanding the
bar codes of purchase items and waving the RFID-enabled credit card near
a reader, resulting in a purchase that is deducted from the card account
(MasterCard Canada, 2007). Paypass has been successful and is being
expanded.
Paypass and Speedpass raise potential ethical spatial issues. For instance,
Paypass insiders at MasterCard could make copies of certain customer records
including their ID codes, and provide the information to outsiders who could
use it for identity theft or covert tracking. RFID technology serves as an
enabling of ethical issues, rather than being the inherent issues. ExxonMobil,
MasterCard Canada, and others providing these services face the ethical issue
of how much to expose customers to such risks, versus pushing the marketing
appeal of an innovation in customer convenience. As tags and readers become
more prevalent, the issue might grow.
An example of high-profile adoption of implanted RFID tags has involved
Mexico’s Office of the Attorney General in Mexico City (Weissert, 2004).
In 2004, the then attorney general of Mexico, Rafael Macedo de La Concha,
along with 160 others in his office, had an implant of an RFID tag from
VeriChip (Applied Digital Solutions, 2007) that enabled him to gain access to
the secure areas of the headquarters office, including the Federal Anti-Crime
Information Center (Weissert, 2004). During the implantation procedure, a
sixteen-digit ID code is scanned by a portable reader (Weissert, 2004). The
intent of the implant policy was to overcome alleged corruption of some
officials. For an RFID-implanted attorney, the reader devices in the head-
quarters can store information on when, where, and by whom confidential
information was accessible. Since kidnapping is a critical problem in Mexico
City and the surrounding area, there is a perception that the implants could
serve as a way for police to track the location of victims, but implanted pas-
sive RFID tags can presently only be read at distances of 100 feet or so,
so the employees would benefit very little. Ironically, the suspected pres-
ence of a tag might expose an implanted employee to greater bodily risk, if
kidnapped.
Having the very-visible Mexican Attorney General’s Office implanted

raises ethical and privacy issues. For instance, employees would be exposed
to the risk of readers being illicitly located outside of the headquarters’ office.
The ethical issue for the attorney general and other leaders is how much to
inform the employees about risks, without compromising the program. From
an ethical theory standpoint, this raises the question of utilitarianism (“ends
justifies the means”).
In summary, following a brief foundation in ethical theory, the section
covered some examples of spatial ethical issues involving business and gov-
ernment. Because of the growth in GIS and spatial technologies, related
ethical questions are likely to expand and take new forms.
PRIVACY
Privacy is defined as the capability of an individual to determine and decide

“when, how, and to what extent information about him/her is communicated
to others” (Westin, 1967, cited in Cho, 2001). The U.S. Supreme Court has
held the right to privacy is provided in the First, Fourth, Fifth, and Ninth
Amendments to the Constitution (Griswold v. Connecticut, 381 U.S. 679
(1965)). At the same time, some states have legislated privacy-protection
rights.
A number of U.S. federal laws (see Table 11.1) determine privacy rights,
in some instances covering locational aspects. The Privacy Act of 1974 pro-
vides the foundation in the U.S. of privacy protection of government data.
Counterbalancing this act are the Freedom of Information Act of 1966 and
subsequent updates, which authorize access to many types of public infor-
mation. For the private sector, a series of acts, starting with the Fair Credit
Reporting Act of 1970, regulate certain industries by giving consumers access
to, and awareness of their credit information.
Two acts specifically give privacy protections for locational information.
The Location Privacy Protection Act of 2001 provides that a company that
has spatial information must notify customers and obtain their permission to
use it. Consent is required to sell the data. The Act does not cover public
spatial information, which is governed by the Privacy Protection Act of 1974
and others. The Wireless Privacy Protection Act of 2003 requires customers’
consent for ISPs or other firms to provide location information from GPS-
enabled wireless devices such as cell phones. The two locational acts are
forward-looking. While they put strong restrictions on what businesses can
do with spatial data of customers without consent, they may be hard to enforce.
They rely a lot on the integrity and ethical propriety of managers and leaders
in business.
The state of privacy regulation today has evolved in major ways from the
nineteenth century view that privacy involved defining a geographic area in
which a person or people can be left alone (Curry, 1999). In a notable Harvard
PRIVACY 337
TABLE 11.1 U.S. Federal Privacy Legislation Significant for GIS

Privacy Act of 1974
Limited protection given on privacy for government databases.
Privacy Protection Act of 1980
Protections regarding privacy for communications and public materials. The materials
cannot be seized, unless there is a criminal offense involved.
Data-Matching Legislation of 1988.
Prevents large dossiers on individuals through record-matching of databases.
Privacy of Financial Information (Fair Credit Reporting Act of 1970; Fair Debt Collection
Practices Act of 1977; Right to Financial Privacy Act of 1978)
Provide regulation of data in credit bureaus. Keep debt organizations from sharing consumer
debt information with 3rd parties. Regulation of access to data in financial institutions.
Transactional Information Privacy (Telecommunications Act of 1996; Cable
Communications Policy Act of 1984). Require cable firms to let customers know what
information they collect and what the uses of it are.
Protect transactional information regarding telephone calls
Children, Video, Health, and Education Privacy Legislation (a group of privacy statutes
1974 to 2002)
Regulation of information and records on identifying children on-line, videotape rentals,
health information, student records, and driver’s license info.
Location Privacy Protection Act of 2001
Firms must notify customers about the presence of location specific collecting of data, and
obtain permission to use the data. The information cannot be sold or used without
consent. Implied is a right to privacy for locational information.
Wireless Privacy Protection Act of 2003. It amends the Communications Act of 1934 “to
require customer consent to the provision of wireless call location information.”
Freedom of Information Act of 1966; (Electronic Freedom of Information Act Amendments
of 1996).
Protects privacy for individuals who “utilize internet-connectable devices that can pinpoint
a person’s location.”
Source: Cho, 2004
Law Review article in 1890 co-authored by future U.S. Supreme Court Justice
Louis Brandeis, the individual’s formal privacy right was defined centering
on the home and its close surroundings as the locus for private activities. This
article was a watershed, after which privacy could be defined legally instead
of solely as an ethical concept.
A particular problem of spatial privacy concerns the perception that GIS
and maps will remain accurate as the user drills down further and further to
larger scales (Curry, 1999). In fact, this drill-down mostly leads to diminished
accuracy, which ironically reduces the need for privacy rights. However,
spatial resolutions are continuing to increase and become available to wider
publics, as manifest in 2005 by the arrival of Google Earth, Microsoft Virtual
Earth, and other web-based, high-resolution map viewers. As everyone gains

access to high-resolution imagery, the very concept of privacy will change,
and privacy legislation might have to be weakened (Curry, 1999).
Public pressure may build up not to release spatial data of the ultimate
resolution. For instance, seeing the behaviors of individuals in their fenced
backyard may raise public outcry. An analogy is what happened in 1996
when Lexis-Nexis was rumored to shortly introduce its “P-Trak System,”
that had a record on nearly every person in the U.S., including the mother’s
maiden name and the individual’s social security number. Facing a public
uproar, Lexis-Nexis decided to release the product without social security
numbers. Google has felt analogous resistance of some foreign governments
and others to its high-resolution map images, and this may portend a growing
spatial-privacy debate. The nature of spatial-privacy has changed, so more
is acceptable to be revealed today about the details of a person’s home or
yard (e.g. from satellite imagery) than occurred in the nineteenth century.
However, there are limits on how far the spatial-privacy concept can be taken
to offer protection, an issue that is central to this chapter’s case study.
Another source of privacy controversy is geodemographics, introduced in
Chapter 10. Geodemographics tends to characterize an entire small area with
one profile, reducing the individual to what is called a “digital individual”
(Agre, 1994) or “digital personae” (Clarke, 1995). In other words, the real
person may be lost, and replaced by a generic digital profile. The individual
might lose control of the virtual self created for him or her (Curry, 1999).
If John lives in the “Trendsetters” small-area but is old-fashioned, for many
purposes affecting him, John is misrepresented. However, he has no control
over it. It may boil down to cutting off at the geocoding stage much of the
geodemographic information. Some point to the need that may arise to re-
strict geocoding through law or regulation (Curry, 1999), in order to restrain
what some consider excessive geocoding that might not only threaten the
individual’s privacy but also misrepresent it, if it is presented in a geodemo-
graphic form without caveats. Excessive geocoding might for instance enable
categorizing buildings or a few households in such detail that unsolicited
target marketing or government micro regulations (e.g. zoning) would be
burdensome.
In short, privacy is an old concept, which has become formalized by leg-
islation. However, the nature of privacy itself is changing. For instance, the
old concept of total privacy of the home is no longer possible through spatial
and other technologies. The challenge is to see today what are the limits on
how far the privacy concept can be reduced. It is an ongoing controversy in
the courts, Congress, and the public arena.
GIS Codes of Ethics recommend ethical approaches for GIS practition-
ers. They have been brought forward by non-profit organizations, including
by the Urban and Regional Information Systems Association (URISA) and
the American Society for Photogrammetry and Remote Sensing (ASPRS).
URISA’s Code of Ethics, adopted in 2003 and seen in Table 11.2, emphasizes
LEGAL ISSUES 339
the GIS professional’s obligations to society, to employers and funders, to

colleagues and the profession, and to individuals in society. Provisions are in-
cluded that avert or reduce some of the ethical and privacy problems covered
in this chapter. The URISA Code includes practicing integrity, providing clear
and accurate information, trying to do what is right (not just legal), protecting
individual privacy especially as regards sensitive information, encouraging
the autonomy of the individual, striving for quality work, and being honest
and professional. For practitioners, the provisions balance the business goals
of efficiency and profit-maximization, with concern that the individual is re-
spected, privacy preserved, and the quality of work and integrity emphasized.
The inclusion of this Code is not meant to imply it is better than other codes or
appropriate for all business situations, but to only offer it as one an example
of a code.
LEGAL ISSUES
Law is “the body of enacted or customary rules recognized by a community

as binding” (Cho, 2005). In contrast to ethics, law tries to be clear-cut on
regulating the actions and behavior of citizens. In the U.S., common law
prevails. It is a widely accepted and general body of law that has evolved
over a long period of time prior to the legislation on it. New laws are added
by (a) legislation and (b) court decisions. Common law in the U.S. has the
major characteristics of “individual rights, equality before the law, and the
universality of the law” (Cho, 2005).
There are many categories of common law that apply to GIS and examples
of spatial problems. They include:
Civil law. Law that regulates the conduct between individuals.

Contracts define relationships between parties in society. Example: de-
termining whether the contract between a company and a GIS out-
sourcing firm has been breached for poor performance. Torts such as
defamation, trespass, liabilities, negligence, result in litigation to de-
termine court actions. Examples: Determining if a GIS data provider
should pay a fine for errors in its commercial data files. Evaluating lia-
bility for supply-chain errors due to a large lot of malfunctioning RFID
tags.
Criminal law. Law that governs charges of crimes of individuals and result-
ing in decisions on punishment. Example: To determine if executives of
a satellite company authorized providing data to a Mafia organization,
which used it to locate targets for destructive attacks.
Procedural law. The law regulating the procedures of the court system.
Technical legal rules that might determine the outcome of the appeal
process of a GIS legal case, such as occurs in this chapter’s case study.
TABLE 11.2 GIS Code of Ethics. Urban and Regional Information Systems
Association. Adopted 4/9/2003
I. Obligations to Society
The GIS professional recognizes the impact of his or her work on society as a whole, on
subgroups of society including geographic or demographic minorities, on future
generations, and inclusive of social, economic, environmental, or technical fields of
endeavor. Obligations to society shall be paramount when there is conflict with other
obligations. Therefore, the GIS professional will:
1. Do the Best Work Possible
Be objective, use due care, and make full use of education and skills.
Practice integrity and not be unduly swayed by the demands of others.
Provide full, clear, and accurate information.
Be aware of consequences, good and bad.
Strive to do what is right, not just what is legal.
2. Contribute to the Community to the Extent Possible, Feasible, and Advisable
Make data and findings widely available.
Strive for broad citizen involvement in problem definition, data identification, analysis,
and decision-making.
Donate services to the community.
3. Speak Out About Issues
Call attention to emerging public issues and identify appropriate responses based on
personal expertise.
Call attention to the unprofessional work of others. First take concerns to those persons;
if satisfaction is not gained and the problems warrant, then additional people and
organizations should be notified.
Admit when a mistake has been made and make corrections where possible.
II. Obligations to Employers and Funders
The GIS professional recognizes that he or she has been hired to deliver needed products
and services. The employer (or funder) expects quality work and professional conduct.
Therefore the GIS professional will:
1. Deliver Quality Work*
2. Have a Professional Relationship*
3. Be Honest in Representations (4 subcategories not shown)
Be forthcoming about any limitations of data, software, assumptions, models, methods,
and analysis.
III. Obligations to Colleagues and the Profession
The GIS professional recognizes the value of being part of a community of other
professionals. Together, we support each other and add to the stature of the field. Therefore,
the GIS professional will:
1. Respect the Work of Others* .
2. Contribute to the Discipline to the Extent Possible*
(continued)
LEGAL ISSUES 341
TABLE 11.2 (continued)
IV. Obligations to Individuals in Society

The GIS professional recognizes the impact of his or her work on individual people and
will strive to avoid harm to them. Therefore, the GIS professional will:
1. Respect Privacy
Protect individual privacy, especially about sensitive information.
Be especially careful with new information discovered about an individual through
GIS-based manipulations (such as geocoding) or the combination of two or more
databases.
2. Respect Individuals
Encourage individual autonomy. For example, allow individuals to withhold consent
from being added to a database, correct information about themselves in a
database, and remove themselves from a database.
Avoid undue intrusions into the lives of individuals.
Be truthful when disclosing information about an individual.
Treat all individuals equally, without regard to race, gender, or other personal
characteristic not related to the task at hand.
∗ Subcategories
not shown
Sources: URISA, GIS Certification Institute (GISCI).
Although this chapter mainly concerns legal matters in the U.S., other
countries’ legal systems also determine GIS cases. International law regulates
the conduct by independent states among themselves (Cho, 2005). It is very
complex with two or more jurisdictions competing, and beyond the chapter’s
scope. However, since spatial technologies apply globally, international law
is relevant for international geobusiness.
Six major areas of GIS and the law are the following (Onsrud, 2001):
1. intellectual property conflicts

2. conflicts that involve personal information privacy
3. liability issues due to damages and losses from GIS and spatial data and
technology
4. conflicts involving competing access to spatial information from gov-
ernment, private sector firms, nonprofits, and individuals
5. conflicts regarding the commodity status of information in spatial
libraries
6. problems and conflicts from the sale of tax-funded spatial data by some
governments, and the restrictions sometimes placed on the use of the
data
In this section, the first four areas are given the most attention. For the first,
the major areas of copyright, trade secret, and patenting are reviewed and
examples given with respect to GIS and spatial phenomena. For the second,
issues of geosurveillance are highlighted.
The third area of liability issues is mostly resolved by contract and warranty
law (Onsrud, 1999). However, if the harm is alleged to have been done to the
general public, then it is governed by tort law. An example of such an issue
involved charter boat captains offshore Alabama (Cho, 2005). Using GPS,
the captains take passengers to “private reefs” that are natural and untouched,
appealing to the typical nature-loving passenger. The legal issue arises when a
charter captain tries to prevent a fisherman from bringing a GPS unit on board
the charter boat. The legal issue to determine is whom the GPS technology is
damaging—the captain or the fisherman.
Issues 4 to 6 revolve around access to spatial information. Can particular
parties benefit or not from the sale of the data; and what restrictions might
parties place on the data? Many of these issues are similar to non-spatial
information legal conflicts and often follow the same precedents in contract
law. However these spatial data issues vary considerably between countries.
Some such as the U.S. favor free provision of government data and others
such as the UK have in the past provided considerable government fees for
data access.
The full set of legal problems on this list and more are reviewed in other
sources (Cho, 2001; Cho, 2005).
Returning to the first legal focus topic of intellectual property law and GIS,
intellectual property is defined as “the property of the mind or intellect [which
is] any intangible thing that gives one an operational and functional advan-
tage over others.” (Cho, 2005). The law of intellectual property protects the
intellectual property of creative people for limited periods of time. It includes
copyright, trade secrets, and patents. In the GIS world, there are many creative
people and organizations that produce a large volume of intellectual property
every month. Some of this property is protected by the developers through
legal means. Later, when the intellectual property gets into widespread use,
disputes can arise about the protection and whether it is sufficient to stop
similar property brought forward by others.
Copyright is a form of protection for creative works. There are two re-
quirements for the works: (1) they are original, and (2) they are fixed in a
durable medium. For instance, new GIS software that is original and in a
lasting storage form can be copyrighted which prevents others from copying
the software without permission of the owner (Johnson, 2001). The copy-
righted software is prohibited from being reproduced, distributed in the form
of copies, or shown publicly, without obtaining permission from its author
or authors (Johnson, 2001). The present term of copyright coverage is the
author’s lifespan plus seventy years. If a user legally obtains copyrighted
material, the statutory exception of fair use may apply, meaning the user can
make copies for personal use, but is prohibited from distributing the ma-
terial. A subtle aspect of copyright is that facts, ideas, and data cannot be
LEGAL ISSUES 343
copyrighted, but the expression of them can—a legal concept called the
“merger doctrine” (Cho, 2005). Hence, the population of a place on a map
cannot be copyrighted, but the map that portrays the population size can.
Several examples of legal cases illustrate the issues with copyright for
GIS and spatial technologies. In Kern River Gas Transmission Co. v. Coastal
Corp (1990), a plaintiff created a map from topographic features of USGS
topographic maps, but added a gas transmission pipeline and data that the
company collected. The court’s ruling was that the company did not have
copyright protection, since the gas elements shown on the map, even though
“new,” were pictorial and hence ideas. The “merger doctrine” applied so the
map could not be copyrighted (Cho, 2005). This points to the need for spatial
companies to go beyond simple pictorial elements on maps and do more
complicated map expression to achieve copyright.
A second example underscores that converting a paper map to digital form,
without changes, does not circumvent the copyright of the paper map. In
Albert R. Sparaco v. Lawler, Matusky, Sekklly Engineers, a surveyor had cre-
ated the original plan in the form of a paper map (Cho, 2005). A new team
came in of owner, architect, developer, and a new surveyor, and copied the
original surveyor’s map into digital form. The court ruled that the original
surveyor had the copyright, and his copyright had been infringed, so license
fees would need to be paid. An implication of this case for the spatial industry
is that digitizing is considered a form of copying, so the rights of the de-
signer of the original source map are preserved. With rampant scanning and
digitizing in place in business and society in general, this case cautions that
copyright is held by the originator. There are other cases and precedents on
copyright of GIS and spatial materials that cover databases, creativity in maps,
software, and other aspects (Cho, 2001, 2005; Moon, 2005) that provide a
basis for businesses to evaluate copyright situations and make decisions on
the strength of protection.
Trade secret provides protection to a company from losing a competitive
advantage involving a key information item (Johnson, 2001). Certain types of
information can be kept secret, that vary between jurisdictions. The following
steps are all required for a trade secret to hold up in court: (1) novelty; (2) con-
stitutes an economic investment for the holder; (3) involved significant effort
to develop; and (4) the company exerted effort in keeping it secret (Johnson,
2001). Trade secrecy laws are jurisdictional and vary in their tests of whether
a trade secret is maintained over time. For GIS and spatial technologies,
novelties keep occurring that might benefit by trade secret protection, so it
would appeal. However, in many cases GIS firms, which tend to be small,
fast-moving, and changing, may not be able to spare the time and tenacity to
achieve and maintain a trade secret.
A patent involves a formal series of steps carried out with the U.S. gov-
ernment to protect an original invention. If a patent is granted, the inventor is
given protection against others from using, making, or selling the property,
while also giving the inventor the ability to license others to use the patented
property (Johnson, 2001). An issue for the prospective patent proposal is that
a lengthy and costly search must be made to determine whether the invention
has already been patented. If parts of it have, then licenses need to be obtained
for those parts. The patent term in the U.S. is either seventeen years after the
date of issue or twenty years from the earliest date of claimed filing, the latter
applying for patents issued since 1995. In 2004, the U.S. Patent and Trade-
mark Office had registered 104 GIS patents, mostly in “image processing,
GPS navigation and routing, vector- and raster-based geographic data, map-
based directory systems, 3-D interactive image and terrain modeling systems,
and methods for mapping and conveying product location” (Cho, 2005). If
there is a patent infringement, then the patent holder can file a lawsuit suit
against the infringer, which must prove that the infringer violated the patent,
often a complex undertaking.
A well-known but controversial patent example is U.S. Patent No.
6,240,360, granted to Sean Phelan of Multimap Inc., a London-based com-
pany, on May 29, 2001, and listed as a computer system to identify local
resources (Reed, 2003; Cho, 2005). It is also covered by European Patent
EP0845124B and patents in other countries. Hence Multimap Inc. can claim
royalties on it. The patent covers a fairly broad class of arrangements for
map serving over the internet (see Figure 11.5). The arrangement involves a
request for a map by a client computer. The request goes to two servers: (a)
an information server that retrieves and responds with data on at least one
place of interest, and (b) a map server that responds to the map request with
a map. The information on the place or places of interest is overlaid on the
requested map, giving the full display to the client. For a patent infringement
to occur, all the steps shown in the figure must be involved.
To successfully challenge this patent, the challenge must be based on
evidence that the patent is either obvious or not new, with “new” referring
to before the date of the U.S. patent application of August 16, 1995. The
patent has been upheld so far and is in effect until 2016. The Multimap
patent dominates in internet mapping, although it doesn’t cover all types of
system arrangements. Some critics (Radcliffe, 2003; Reed, 2005) question
the validity of this patent, not considering it to be “new” at the time of its
granting in 1995.
In summary, patents require a lot of effort to achieve and often are outmoded
by fast-moving technology. However, some in the spatial industry such as
Multimap have created a durable and competitive patenting.
The discussion now turns to the second legal highlighted topic of geo-
surveillance, which exemplifies issues in information privacy law. Geosurveil-
lance refers to using spatial technologies (often satellites, sensors, GPS, or
RFID, combined with GIS) to monitor the activities of individuals. This was
already illustrated in the privacy section with respect to monitoring of the
very small but growing number of persons implanted with RFID devices.
A less intrusive form is to provide wearable RFID tags or GPS monitors
LEGAL ISSUES 345
Figure 11.5 Design Covered by U.S. Patent No. 6,240,360, “A Computer System
to Identify Local Resources”. Source: Cho, 2005
for classes of persons, such as children in school, offenders on parole, or

soldiers in combat. For instance, in Iowa, some criminal offenders are out-
fitted with a wearable device, iSecureTrac, that is GPS-enabled and trans-
mits the offender’s location to a map server, which allows parole officers to
monitor movements of the parolee (Chabrow, 2002). The map can be con-
figured for a specific parolee and show obstacles that must be avoided, e.g.
a paroled sex offender must avoid a school playground. Such monitoring
devices have become more common with children as well, so parents or
day-care workers can see at all times where children are located and avert
any dangers.
Another type of geosurveillance is satellite mapping (Monmonier, 2002),
introduced earlier in the book. It can be used by businesses, law enforcement,
and intelligence services to monitor changes to the earth, including land fea-
tures, built structures, transportation, agriculture, and military movements.
For instance, in attempts by law enforcement units to destroy marijuana
crops, investigators used a GIS assisted by an expert system to interpolate
and combine information, enabling it to achieve an accuracy of 87 percent
for ninety sites (Monmonier, 2002). However, since police were now able
to locate nearly all the sites, the illegal growers responded by bringing the
marijuana plants indoors and using hydroponic methods. Such a sequence
of events points to the legal issues of conflicts involving privacy versus geo-
surveillance. The chapter case study will pick up on the legal issues involved
in satellite sensing of alleged criminal activity.
In summary, geosurveillance, whether by RFID, GPS, satellite, or other
technology, has benefits and costs to the parties involved. Not only are there
ethical issues, but legal disputes as well. As new technologies appear and more
cases come forward, the laws of privacy related to geosurveillance would be

refined.
CASE STUDY OF LEGAL ASPECTS OF SURVEILLANCE:

KYLLO V. UNITED STATES
Before examining the Kyllo case, it is necessary to set a backdrop on rulings

regarding remote imaging and sensing of property. In a key case, Dow Chemi-
cal Company v. United States in 1986 (Monmonier, 2002), the Environmental
Protection Agency (EPA) contracted with an airborne photogrammetry firm
to take overflight photos of Dow Chemical’s huge plant in Midland, Michigan,
which revealed illegal activities taking place. The photos not only revealed
buildings, but exposed equipment, conduits, and fittings between the build-
ings. Dow in mounting its defense argued that the interior areas between
buildings were equivalent to the private courtyard and walled areas of homes,
referred to as curtilage. The Court of Appeals ruled in favor of Dow, but it
was overruled in favor of the EPA by the U.S. Supreme Court—that the open
field areas between plant buildings have much less privacy than the curtilage
of homes, and thus the concept of private home was not allowed (Cho, 2005;
Bruzzese, 2006) This resulted in a concept of “intimate details,” i.e. that a
home’s curtilage could reveal intimate details of private personal life, but not
so for outdoor areas of the industrial plant.
With this as backdrop, the 1998 case of United States v. Kyllo involved the
legality of police obtaining a search warrant for the home of Danny Lee Kyllo
(Woessner and Sims, 2003; Bruzzese, 2006). Police suspected that Danny
was growing marijuana in his triplex home. They used ground-based FLIR
imaging to show unusual thermal activity from the home. Forward looking
infrared technology (FLIR), sometimes called thermal imaging, is a sensor
technology that can detect heat being emitted from distant objects. The police
used this as evidence to obtain a search warrant, and indeed discovered a
marijuana growing operation taking place in the heated environment inside.
The lower court ruled in favor of Mr. Kyllo, reasoning that “intimate details”
such as baths, showers, and dryers, could be revealed by thermal sensing
in homes throughout the country (Monmonier, 2005). However, a year later
the U.S. 9th Circuit Court of Appeals overruled the argument that the thermal
technology did not reveal much about the inside of homes. The Supreme Court
in 2001 reversed this decision in Mr. Kyllo’s favor. The justices argued that
without a search warrant, surveillance use of a technology not in common use
violated the normal expectation of home privacy (Bruzzese, 2006). Although
there was a division with the minority doubting that thermal imaging revealed
anything intimate, most agreed it did. However, they left the door open for
a change of interpretation if thermal imaging technology were to become
commonplace. The implication is if it were commonplace, the concept of
privacy would change to expect this surveillance as an ordinary police activity.
CRIME AND SECURITY 347
Today, as surveillance technologies of different types become both more

prevalent and commonplace, the Kyllo case is a key ruling that touches on the
fundamentals of what is privacy and how far can technology go in invading
and reducing it. Spatial technologies particularly sensing and imaging have
the potential to reduce privacy, in the home, workplace, or elsewhere. Kyllo
is not the last word on these surveillance and privacy issues, but encourages
awareness of the issues and risks.
CRIME AND SECURITY
GIS and spatial technologies are a valuable asset for businesses and need to
be protected through security procedures. If they are abused, damaged, or in-
truded into, the results can be injurious to businesses and organizations. Some
of the spatial crime and security issues considered in this section are shared
in common with information systems. Consider that many key functions of
the book’s case-study businesses would be interrupted, stopped, or rendered
inaccurate by security breaches. For firms such as Rand McNally and Sears
that have real-time dependencies of spatial technologies from hour-to-hour,
protracted damage could be devastating. Another aspect of this section is that
spatial technologies can be employed to help criminals.
Consider that financial theft can occur when hackers break into databases,
which may include spatial information. This can be extended further to iden-
tity theft. For instance, the databases of Equifax Canada, a branch of the huge
credit information firm, were broken into, and over 1,400 personal records
stolen, with financial, credit, and personal data including addresses, which
could be geocoded for spatial analysis. The criminals who were later appre-
hended reported their interest in the spatial attributes (Cho, 2005).
It is well-known that drug smugglers and illegal distributors use GPS
technologies to pinpoint transit routes and drop-off points. This parallels the
U.S. military uses of GPS and GIS for supply chain in Chapter 4.
Perhaps the most notorious criminal use of spatial technologies was the
likely use of portable GPS equipment by the 9/11 terrorists. Although this
cannot be definitively proven, much circumstantial evidence points to it. One
of the 9/11 terrorists had tried to purchase four GPS units at a shop for
pilots in Miami three weeks before the attack—he ended up purchasing one
(National Commission on Terrorist Attacks, 2004). Zacarias Moussaoui, the
alleged twentieth hijacker, tried to purchase a GPS unit, according to U.S.
prosecutors (Smith, 2002). Some reports have indicated that Mohamed Atta,
leader of the attacks, visited the observation deck of the Twin Towers on
September 10 in order to confirm the spatial coordinates of the towers and
enter them into the GPS units used the next day on the two flights bound for
the Twin Towers (Smith, 2002; CNN, 2002; Cooperative Research History
Commons, 2007). It’s likely that the terrorist pilots did not know how to
operate the sophisticated planes’ guidance systems, so relied instead on these

handheld GPS devices.
Regardless of exactly what happened, this example emphasizes that spatial
technologies have a potentially evil and sordid side, involving major crimes
and even terrorism. It underscores the need for GIS security.
GIS security involves steps that are commonly utilized in IT security.
This includes physical security, passwords, audit trails, encryption, software
maintenance, training of users, and scrutiny in hiring practices for GIS and
IT professionals (Oliva, 2005; Dhillon, 2007). As pointed out by Dhillon
(2007) often the greatest vulnerability of security systems are at the very
core—involving the trusted people in the GIS and IT departments. At the
higher levels of these departments and the firm as a whole, the potential
security threat becomes that much greater, since the higher-ups have access
to most, or all of the firm’s information (Oliva, 2004).
Dhillon (2007) considers the security layers to be like layers in an onion The
technical system is the core of an organization’s security, surrounded by the
formal system of security controls, which in turn is surrounded by the informal
system of security controls. The technical system applied to GIS and spatial
technologies shown in Table 11.3 consists of supportive controls, preventative
TABLE 11.3 Technical Security Controls for GIS and Spatial Technologies
Supportive controls.
User identification. Achieved by required and options access control through passwords
and other means.
Cryptography. Encryption of messages and information.
Technical security administration. Covers IS and GIS areas.
System protections. Software and system design that reveals security threats or breaches,
such as modular programming, separation of business processes, and reuse of objects in
software.
Preventive controls
Authentication. Smart cards, tokens.
Authorization. Authorizing users, IS and GIS employees, and others.
Access control enforcement. Access control lists, user profiles, clear roles.
Protected communication. Virtual private network, packet sniffing, wiretapping.
Transaction privacy.
Detection and recovery
Audit. Needs to extend to full set of GIS and spatial technologies.
Detection of intrusions and containment of problems, damages, and outbreaks.
Restoration to secure state.
Virus detection.
Source: Modified from Dhillon, 2007

CRIME AND SECURITY 349
TABLE 11.4 Formal and Informal Security Controls for GIS and Spatial
Technologies
Formal Controls Informal Controls
Preventative
Security-responsibility allocation Security awareness program

Personnel controls – separation of Security training including technical and
responsibilities and duties organizational issues
Clarity of security roles between IS and Development of security subculture
GIS Departments throughout the organization
Detection
Personnel background checks Encourage informal feedback

Clearances, rotation of duties Reward structures for reporting breaches
Management audits
Continual risk management
Recovery
Disaster and business continuity plans Provision of ownership of activities

Incident-response capability Encouraging sense of stewardship for
recovery
Source: Dhillon, 2007.
controls, and detection and recovery (Oliva, 2005; Dhillon, 2007). Since GIS
and spatial technologies often have coupled technologies that work together,
such as GIS and GPS, the technical security controls need to encompass the
full array of spatial technologies, even if they reside in different departments.
Formal controls are those formal rules in the organizational system that
support the technical structure of security. They mitigate the misunderstanding
of rules and assist in determining clear roles and responsibilities. As seen in
Table 11.4, formal security controls emphasize allocation of responsibility
for security and separation of duties of security personnel. Many of the case-
study businesses in this book have separate IS and GIS departments. Formal
controls need to be set up to formalize what are the security responsibilities
between the two. The GIS department may be less versed in such controls,
since its people tend to be trained in geography, environmental, and other areas
outside business and IS. Informal controls, which also are organizational, help
in stimulating awareness of security as an issue through training, education,
and informal user support and encouragement. Informal controls (Dhillon,
2007) are shown in Table 11.4. They involve in general creating an atmosphere
that recognizes organization-wide security awareness, training, and employee
“ownership” of GIS and information security. Many tenets of the GIS Code of
Ethics in Table 11.2 encourage employees to take due care in implementing
security to forestall forseeable misuses of spatial technologies.
GIS security needs to involve people who have had training and education
in it. This includes broad-based awareness training across the organization.
The users and employees need to understand what are the underlying rea-
sons it is important and what are the broad strategies to achieve it. At the
level of departments that interact with GIS and spatial technologies, training
needs to focus on specific security threats and preventions. For example, if
an operations department manages a corporate supply chain that is RFID en-
abled, those employees need to be aware of the potential for counterfeit tags,
improper tagging, and intrusion into the storage files connected to supply
chain items. For security and networking managers and staff who deal with
GIS and spatial technologies, they need continual, in-depth, and up-to-date
training in order to fulfill their challenging roles related to management and
effecting the controls in Tables 11.3 and 11.4, as well as to keep up with the
emerging technologies that have characterized this field.
GIS and spatial security may further need to meet the provisions of fed-
eral laws on information such as the Sarbanes-Oxley Act of 2002, HIPAA
(Health Insurance Portability and Accountability Act of 1996), and Gramm-
Leach-Bliley Act of 1999. The Sarbanes-Oxley Act has strict provisions that
guard the integrity and correctness of internal financial information in busi-
nesses. It requires internal controls on financial data that include information
systems security. It applies to the spatial data on finance for example in the
book’s banking and insurance case-study firms. HIPAA mandates rules and
procedures to maintaining security and privacy of health-care information,
for simplifying health-care accounting, and for the secure exchange of infor-
mation among health-care firms. It would apply for health-care firms such
as Kaiser Permanente and Baystate. The Gramm-Leach-Bliley Act has pro-
visions to assure secure, accurate, and confidential customer information. It
limits the use of customer information in spatial applications that are available
publicly.
In summary, achieving GIS and spatial security is a daunting task that
every spatially based business and organization needs to set as a goal. It is
a continuing and never-ending battle of security readiness and capability,
versus new emerging threats (Oliva, 2005; Dhillon, 2007). The GIS security
can have great technical barriers and controls in place, but in the end it is
no better than its “trusted” agents. Some of the worst security breaches have
occurred from trusted security professionals on the inside. This section has
stressed the various levels of security controls that are not only technical but
managerial and organizational.
CHAPTER SUMMARY
Spatial technologies have developed extensively in business and serve im-

portant functions. As their capabilities have improved, the benefits that are
REFERENCES 351
sought can sometimes be misdirected or usurped causing harm and dam-

ages. Ethical questions arise regarding how much businesses should push
their profit-enhancing potential versus fairness and respect for stakeholder
individuals. At its worst, abuse can take the form of malicious activities,
crime, invasion of privacy, financial harm, and information sharing. The legal
system is coping with defining the limitations and boundaries of proper con-
duct regarding GIS, geographic information, and spatial technologies. Busi-
nesses can strengthen the threat of misuse, intrusions, and crime by adopting
GIS security plans and putting in place a skilled and honest team to effect
controls and preventatives. Achieving and maintaining such security requires
continual effort, education, and quick adaptation to new technologies.
REFERENCES
Agre, P.E. 1994. “Understanding the Digital Individual.” The Information Society,
10: 73–76.
Ahearn, Sean C. 2004. “Case Study 3: GIS at the World Trade Center After Septem-
ber 11, 2001.” Pp. 241–252 in Clarke, Keith, Getting Started with Geographic
Information Systems, Upper Saddle River, NJ: Prentice Hall.
Barr, Robert. 1997. “Nowhere to Hide.” The Geographical Magazine, 69: 30–31.
Biever, Celeste. 2004. “RFID Chips Watch Grandma Brush Teeth.” NewScientist.com,
March 17.
Bouchoux, Deborah E. 2005. Intellectual Property: The Law of Trademarks, Copy-
right, Patents, and Trade Secrets, 2nd Edition. New York: Thompson.
Bradley, Theresa. 2005. “You’ve Been Chipped: Microchips Tag People Un-
der the Skin.” Columbia News Service, April 19. Available at http://jscms.jrn.
columbia.edu.
Bruzzese Jr., Joseph. 2006. “Legal Aspects of Forward-Looking Infrared Technology:
Understanding Probable Cause.” Air Beat Magazine: Journal of the Airborne Law
Enforcement Association, October. Available at www.alea.org.
Butler, Al. 2005. “Map Scale: A Guide to Practicing Ethical GIS.” Geoplace.com,
March 1. Available at www.geoplace.com.
Candiotti, Susan. 2002. “Records Suggest Atta in NY on Sept. 10.” CNN.com, May
22.
Chabrow, E. 2002. “Every Move You Make, Every Breath You Talk.” Information
Week, August 30. (Monitoring of offenders.)
Cho, George. 2001. Geographic Information Systems and the Law. Chichester, Eng-
land: John Wiley and Sons.
Cho, George. 2005. Geographic Information Science: Mastering the Legal Issues.
Hoboken, NJ: John Wiley and Sons.
Clarke, Keith. 2004. Getting Started with Geographic Information Systems. Upper
Clarke, R. 1994. “The Digital Persona and Its Application to Data Surveillance.” The
Information Society, 10: 77–94.
Coleman, Kevin. 2003. “Technology and Homeland Security.” Directions Magazine,

March 6. www.directionsmag.com.
Crampton, Jeremy W. 2003. “Cartographic Rationality and the Politics of Geosurveil-
lance and Security.” Cartography and Geographic Information Science, 30(2):
135–148.
Cranor, Lorrie F. 2001. “The Role of Privacy Enhancing Technologies.” Available at
http://www.cdt.org/privacy/ccp/roleoftechnology1.shtml.
Curry, M.R. 1998. Digital Places. Living with Geographic Information Technologies.
London: Routledge.
Curry, M.R. 1999. “On the Possibility of Democracy in a Geocoded World.” Social
Science Computer Review, 17: 10–15.
Curry, M.R. 1999. “Rethinking Privacy in a Geocoded World.” In Longley, Paul
A., Michael F. Goodchild, David J. Maguire, and David W. Rhind, Geographical
Information Systems, New York: John Wiley and Sons, 757–785.
Curry, M.R. 1997. “The Digital Individual and the Private Realm.” Annals of the
American Association of Geographers, 87: 681–699.
Cutter, Susan L., Douglas B. Richardson, and Thomas J. Wilbanks (Eds.). 2003. The
Geographical Dimensions of Terrorism. New York: Routledge.
Wiley and Sons.
Dobson, Jerome E. 2004. “The GIS Revolution in Science and Society.” In Brunn,
Stanley D., Susan L. Cutter, and J.W. Harrington Jr., Geography and Technology,
Dordrecht, The Netherlands, Kluwer Academic Publishers, 573–587.
Durocher, J.M. 2002. “Webrasks CEO Proposes ‘Laws of LBS.’” http://www.
mtravel.com/20614.shtml.
Edson, Curtis, B. Garcia, J. Hantman, N. Hartz, H. Jensen, J. Leale, K. Lewelling, J.
Marks, J. Maxted, B. Moore, B.V. Rivera, and A. Weitzel. 2001. “Code of Ethics
for GIS Professionals.” Madison: Environmental Monitoring Program, Institute
for Environmental Studies, University of Wisconsin Madison. May 8.
EPIC. “EPIC Online Guide to Practical Privacy Tools.” Electronic Privacy Informa-
tion Centre. Available at http://www.epic.org/privacy/tools.html.
Federal Reserve Bank of New York. 2004. “Stored Value Cards: An Alternative for
the Unbanked?” July. New York: Federal Reserve Bank of New York. Available
at 7/2005 http://www.ny.frb.org/regional/stored value cards.html.
Goss, Jon. 1995. “We Know Who You Are and We Know Where You Live: The
Instrumental Rationality of Geodemographic Systems.” Economic Geography,
71(2): 171–198.
Haque, Akhlaque. 2003. “Information Technology, GIS, and Democratic Values: Eth-
ical Implications for IT Professionals in Public Service.” Ethics and Information
Technology Dordrecht, 5(1): 39. Available on Proquest.
Havenstein, Heather. 2005. “Med School CIO Tests RFID for Patients.” Computer-
world, February 11.
Johnson, Deborah G. 2001. Computer Ethics, 3rd Edition. Upper Saddle River, NJ:
Prentice Hall.
REFERENCES 353
Kwan, Mei-Po. 1999. “Gender and Individual Access to Urban Opportunities:

A Study Using Space-Time Measures.” The Professional Geographer, 51(2):
210–227.
Kwan, Mei-Po. 2002. “Introduction: Feminist Geography and GIS.” Gender, Place,
and Culture, 9(3): 261–262.
Kwan, Mei-Po. 2002. “Is GIS for Women? Reflections on the Critical Discourse in
the 1990s.” Gender, Place, and Culture, 9(3): 271–279.
Kwan, Mei-Po. 2002. “Feminist Visualization: Re-envisioning GIS as a Method in
Feminist Geographic Research.” Annals of the American Association of Geogra-
phers, 92(4): 645–661.
Monmonier, Mark. 2002. Spying with Maps. Chicago: University of Chicago Press.
Moon, George. 2005. “Software Patents Cause Mixed Emotions in GIS.” Geo-
place.com, February 8. Available at www.geoplace.com.
Murphy, Clare. 2003. “Tracking Down Your Child.” BBC News, October 28.
National Commission on Terrorist Attacks. 2004. “The 9/11 Commission Report:
Final Report of the National Commission on Terrorist Attacks Upon the United
States.” Washington, D.C.: National Commission on Terrorist Attacks, United
States Government.
Oliva, Lawrence. 2004. Information Technology Security. Hershey, PA: CyberTech
Publishing.
Onsrud, H.J. 1999. “Liability in the Use of GIS and Geographical Datasets.” In
Longley, Paul A., Michael F. Goodchild, David J. Maguire, and David W. Rhind,
Geographical Information Systems, New York: John Wiley and Sons, 643–652.
Pickles, J. 1999. “Arguments, Debates, and Dialogues: the GIS-Social Theory Debate
and the Concern for Alternatives.” In Longley, Paul A., Michael F. Goodchild,
David J. Maguire, and David W. Rhind, Geographical Information Systems, New
York: John Wiley and Sons, 49–60.
Radcliffe, J. 2003. “Death of Copyright—Long Live Patents.” Cambridge Conference
2005, Ordnance Survey. UK Paper 4D.2B. Southampton, England.
Reed, Carl. 2003. “Intellectual Property, Patents, and Web Mapping.” ISPRS Journal
of Photogrammetry and Remote Sensing, 8(3): 33–35.
Scheeres, Julia. 2003. “When Cash is Only Skin Deep.” Wired News. Available at
www.wired.com.
Smith, Greg B. 2002. “Hijacker in City Sept. 10 Used Navigation Tool to Pinpoint
WTC Site.” New York Daily News, May 22.
Smith, N. 1992. “Real Wars, Theory Wars.” Progress in Human Geography, 16(2):
257–271.
Spinello, Richard A. 2003. Case Studies in Information Technology Ethics, 2nd
Edition. Upper Saddle River, NJ: Prentice Hall.
Sui, Daniel. 2004. “Are Robotics Laws Applicable to Location-Based Services.”
Geoplace.com.
The Economist. 2003. “The Revenge of Geography.” The Economist, March 15,
366(8315): 19–22.
Timander, Linda M., and Sara McLafferty. 1998. “Breast Cancer in West Islip, NY:
A Spatial Clustering Analysis with Covariates.” Social Science and Medicine,

46(12): 1623–1635.
Valentine, Gill. 2004. “Geography and Ethics: Questions of Considerability and
Activism in Environmental Ethics.” Progress in Human Geography, 28:2:
258–263.
Warren, S., and L.D. Brandeis. 1890. “The Right of Privacy.” Harvard Law Review,
4: 190–230.
Weiner, Daniel, Trevor M. Harris, and William J. Craig. 2002. “Community Partic-
ipation and Geographic Information Systems.” In Craig, William J., Trevor M.
Harris, and Daniel Weiner, Community Participation and Geographic Information
Systems, London: Taylor and Francis.
Weissert, Will. 2004. “Microchips Implanted in Mexican Officials.” MSNBC.com,
July 14.
Westin, Alan. 1967. Privacy and Freedom. New York: Athenaeum.
Woessner, Matthew C., and Barbara Sims. 2003. “Technological Innovation and
the Appplication of the Fourth Amendment.” Journal of Contemporary Criminal
Justice, 19(2): 224–238.
CHAPTER 12
GIS AND BUSINESS STRATEGY
GIS is one of the many tools and technologies that can be used competi-
tively by business. This chapter evaluates the strategic and competitive role
of GIS. Factors that predispose GIS to rise up in a firm competitively, among
so many other components in business, are considered. Strategy theories
are examined from the marketing and IS fields, including Michael Porter’s
theory of internet strategies and the IT strategic alignment model. GIS prac-
titioners (Huxhold and Levinsohn, 1995) have added useful insights and
frameworks.
The case study of Norwich Union, which was highlighted in Chapter 7
from the standpoint of cost-benefit analysis, is viewed in this chapter from
the different perspective of GIS’s influence on competitiveness and barriers
the firm has encountered in incorporating GIS into its strategy.
Based on the strategy findings for the twenty research case studies in
the book, the chapter develops an evolutionary model of GIS strategy. It
incorporates the dimensions of extent the firm’s GIS is customer-facing versus
internal-facing, extent that geography is important to business revenues and
profits, and degree that a firm has adopted the web-integration enterprise
platform for its GIS, versus the more traditional GIS approaches such as
client server or desktop. Based on the case sample, the latter two dimensions
influence the GIS strategic importance. This model is not quantitative, but
may be useful to managers in making a first assessment of how strategic GIS
can be for an organization.
The chapter next turns to the Rand McNally case, which demonstrates
how the famous map company achieved a high strategic level for GIS. It
355
356 GIS AND BUSINESS STRATEGY
asks what are some of the secrets of its success. The chapter recommends
practical steps for managers to clarify, align, and strengthen GIS strategy. The
chapter finishes by considering future trends influencing spatial strategies and
a chapter summary.
STRATEGIC VISION FOR GIS AND ESTABLISHING

A STRATEGIC PLAN
For business success stories of GIS such as Sears Roebuck and Rand Mc-
Nally, having all the right technologies, skilled people, and a budget to invest
were not enough. Attaining industry leadership with spatial technologies also
requires the vision to foresee years into the future to a spatially-enabled busi-
ness with GIS and associated technologies providing sustained efficiency and
productivity that add value. This points to leaders in these firms who fostered
or developed the vision, gained commitment of stakeholders, and led in mak-
ing it happen through implementation of the strategies over many years. The
intangible leadership factors are crucial, and a part of this chapter.
At some point for most firms, the vision needs to be formalized into the
strategic planning process. Mid-size and large firms are more likely to have
formal corporate strategic plans than smaller ones. The corporate strategic
plan includes the firm’s mission, guiding objectives, mid-term milestones to
reach the objectives, sub-plan for development of employees, and section
on how stakeholders have been involved in establishing the strategic plan
(Tomlinson, 2003; Applegate et al., 2007). The firm needs to have leaders in
GIS who will take the initiative to formulate strategies and gather the support
of company top leadership to include GIS in its business strategies. The GIS
plan needs to be in synchrony with the corporate strategic plan.
It is a breakthrough point at which top leadership understands that GIS is
strategic for the organization (Tomlinson, 2003). Once in the corporate plan,
there needs to be continual effort to monitor progress and keep the planning
components moving forward. The chapter will include case studies of strategic
planning at Norwich Union and Rand McNally that tell contrasting stories.
For Norwich, although GIS has made valuable contributions and provided
some tactical benefits, there hasn’t been the critical mass to establish GIS
as part of Norwich’s overall strategy. On the other hand, for Rand McNally,
there has been strategic vision for GIS and it is not only in the firm’s strategic
plan but GIS is viewed throughout the company as strategic.
THEORIES OF IT ALIGNMENT, INTERNET STRATEGY

AND SITUATIONAL ANALYSIS
This section reviews theories of business strategy that are useful in evaluating
the strategic level of GIS and spatial technologies for firms and industries.
THEORIES OF IT ALIGNMENT, INTERNET STRATEGY AND SITUATIONAL ANALYSIS 357
Most of the theories were developed for information technology, rather than
GIS. The reason is that information technology has had major presence in
business for much longer than GIS, and has much larger and higher profile
than GIS in most firms.
After a brief review from Chapter 1 of strategic grid theory and its relevance
as well as limitations, the theories of IT alignment, internet strategy, and
situational analysis are presented. The relevance of each theory to spatial
technologies is brought out. This section provides the background for the
book’s strategic theory for GIS which is covered later on.
In Chapter 1, the strategic grid was presented (see Figure 1.15) (McFarlan,
1984; Applegate, 2007). The grid considers roles of IT based on four quadrants
consisting of high and low values of two variables, (1) impact of IT on business
operations and (2) impact of IT on strategy. As seen in Figure 1.15, there are
four quadrants based on high and low values for the two variables. In applying
it to GIS, GIS can be substituted for IT in the grid, which leads to the following
four quadrants: (1) support, i.e. GIS has low impact on business operations and
business strategy, (2) factory, i.e. GIS has high impact on business operations
and low on strategy, (3) turnaround, i.e. GIS has high impact on business
strategy and low on operations, and (4) strategic, i.e. GIS has high impact on
operations and strategy.
The strategic grid theory is helpful in gauging how operational GIS is, and
in trying to determine when its impacts rise above the level of operations. For
Sears, GIS is clearly operationally important, but has also been identified as a
strategic factor. It falls into the “strategic” quadrant. At the other extreme the
Large Credit Bank has not put GIS to use operationally, except for inheriting
some operational uses in an acquired bank; nor has it recognized any busi-
ness strategic value. It falls into the “support” quadrant. Each of the book’s
research case studies could be classified in the GIS strategic grid. However,
the usefulness of this theory is limited because it does not recognize the key
factor today in spatial technologies of the internet and doesn’t have a way
that the interactions of GIS and IT can be considered.
IT Alignment
In considering GIS strategies, it is useful to look at models of strategic IS
alignment (Henderson and Venkatraman, 1992; Papp, 2001; Applegate et al.,
2007; Cegielski, 2005), competitive forces (Porter, 2001), strategy and the
internet/web (Porter, 2001; Hagel and Brown, 2001), and situational analysis
(Huxhold and Levinsohn, 1995).
The IT strategic alignment model (Henderson and Venkatraman, 1992;
Papp, 1995; Papp, 2001; Applegate et al., 2007) divides strategy into four
quadrants—business strategy, IT strategy, organizational infrastructure, and
IT infrastructure (see Figure 12.1). Relationships between these quadrants
determine the extent to which business and IT strategies and infrastruc-
ture operate in synergy. The model postulates the closer the alignment that
Figure 12.1 Business-IT Strategic Alignment Model. Copyright 2001, IGI Global,
www.igi-pub.com, reprinted by permission of the publisher.
exists between the quadrants, the greater the synergy. In aligning business
and IT capabilities, for example the business administrative structure and IT
infrastructure need to be consistent (Papp, 2001). Likewise the strategies and
infrastructures need to be consistent, for instance IT strategy may call for
a spatially-enabled supply chain but the IT infrastructure (people, expertise,
networks, RFID equipment) may not be sufficient to support it. A firm’s
strengths and weaknesses in the quadrants can help determine where invest-
ment needs to be prioritized and what results can be expected (Henderson and
Venkatraman, 1992; Papp, 2001).
The IT alignment model can be extended to include GIS Strategy and
GIS Infrastructure (Figure 12.2). Now the alignment needs to exist between
six cells, which requires more time and resources in planning, coordination,
and communication. If the IT and GIS functions and organizational units are
combined together, then the functional integration becomes simpler, again
with only four quadrants, and similar to Figure 12.1. However, based on the
cases in the book, the GIS and IT functions are more likely to be separate,
with loose connections. Regardless of the arrangement, this theory stresses
THEORIES OF IT ALIGNMENT, INTERNET STRATEGY AND SITUATIONAL ANALYSIS 359
ORGANIZATIONAL GIS INFRASTRUCTURE IT INFRASTRUCTURE

INFRASTRUCTURE
Figure 12.2 Business-GIS-IT Strategic Alignment Model. Source: Modified from

Papp, 2001
that GIS strategy can succeed only if effort is made to align it with business
and IT strategies, as well as have a fit between GIS strategy and infrastructure.
A version of the IT alignment model was tested for emerging information
technologies (EITs) in an interview study of senior IT executives (Cegielski
et al., 2005). Both business alignment issues and technical alignment issues
were important for successful and timely IT strategy. Adhering to IT alignment
strategies to implement EITs has the advantage to a firm of being aligned early
on the technology adoption curve, so strategies can last longer (Cegielski et
al., 2005; Luftman, 2001). For an emerging spatial technology, this study
supports early effort to align the new technology with the business strategy,
so as to reap longer and more profitable benefits. Early alignment involves
risk-taking (Luftman, 2001). A book case example of early alignment paying
off is the Motion-Based case, where GPS-enabled devices for runners were
aligned early with the business strategy of athletic performance monitoring.
The early alignment continued to work well in growing its market, leading to
buyout by a larger firm.
Internet Strategy
The internet has profoundly influenced business strategy (Porter, 2001). At
the industry level, the internet has tended to increase competitive challenge
because of more openness of business activities, reduced bargaining power
of suppliers and buyers, lowered barriers to entry, and expanded geographic
range. The internet can strengthen a firm’s strategic positioning by build-

ing on and complementing the firm’s basic business strengths, rather than
being set up as an isolated corporate unit restricted to web services or e-
business (Porter, 2001). This also implies that “old economy” firms can have
the strongest internet strategies by complementing already major or dom-
inant strengths. Effective internet strategies for firms don’t take advantage
of “giveaways” of goods to temporarily bolster revenues and unstable cost
reductions but complement and strengthen existing strategies (Porter, 2001).
This theory implies that the best strategy for spatial internet and map web ser-
vices is to complement existing business strengths, rather than placing them
in an isolated unit. A good case-study example is Sperry Van Ness, which
complemented the existing marketing strength of its regional sales managers
with user-friendly and corporate-based spatial web services on-the-spot in the
regions, rather than placing those services in an isolated unit.
The adoption of the web-services architecture yields many benefits and
increases strategic competitiveness (Hagel and Brown, 2001). Since it requires
standard web protocols and open architecture, a business system can be
quickly and flexibly re-configured by combining internal and external web
services. It encourages firms to move away from secretive, proprietary IT that
depends only on internal staffs toward combining the “best of breed” of web
services which are publicly available. This approach might evolve as far as
a “service grid” or “grid computing,” for example groups of firms that share
parts of their web services (Hagel and Brown, 2001).
This strategy is consistent with the trend for spatial technologies in busi-
ness to grow beyond traditional GIS systems to enterprise-wide systems that
run on web-based architectures (Sonnen and Morris 2005; Guerrero, 2005).
Although it has powerful functionality, traditional client-server GIS requires
considerable geographic and GIS expertise to develop. Often traditional sys-
tems have a limited and specialized set of clients who are knowledgeable
in geography and GIS. The enterprise-wide web-service systems run more
flexibly, with a broader user base, and in an interconnected manner discussed
in Chapter 5.
Situational Analysis
Another strategic method available for GIS strategic practitioners is situ-
ational analysis (Huxhold and Levinsohn, 1995), which is defined as “an
understanding of the organization: its purpose, how it operates, its culture,
and the management style” (Huxhold and Levinsohn, 1995). The elements to
be included in a situation analysis for GIS include the following (Huxhold
and Levinsohn, 1995):
r Fundamentals of the organization

r Management philosophy and style
CASE STUDY: NORWICH UNION 361
r The culture or cultures of the organization

r Driving force for GIS. (Who is pushing for GIS? Impacts of different
driving forces on the project.)
r Technological maturity
r Available resources
r Complexity of the business functions of the organization
r Assessment of implementation success risk/uncertainty. In particu-
lar will there be adequate ROI that does not imply large structural
changes.
The situation analysis adds to the “softer” parts of the background informa-
tion for the strategic plan. Questions need to be asked about the relationship
between the organization and GIS, including what kind of GIS would work
best, what is the human and technological readiness for GIS, how quickly can
the organization learn about GIS, who is its champion, and what is an appropri-
ate and manageable rate of implementation (Huxhold and Levinsohn, 2003).
In addition to the situation analysis, solid information is added such as GIS
hard costs, productivity estimates, and description of existing organizational
structure before developing the strategic plan. These authors see GIS strategy
not as a separate planning process, but integrated with corporate strategic
planning. They even envisage GIS inter-organizational strategic planning,
which they term “multiparticipant” (Huxhold and Levinsohn, 1995). Several
partner organizations would share their strategic planning for GIS. This was
seen in the Enmax case, in which a city-owned utility shared and cooperated
on GIS strategic planning with other local government agencies and utilities.
Their practitioner recommendations support a chapter theme of the necessity
for alignment or even integration of corporate and GIS strategic planning.
Their interest in situational analysis of the softer side of GIS strategy links
with many of the case-study findings.
In sum, several leading models of business-IT alignment and internet strat-
egy, and situational analysis extend the theoretical foundation to understand
and formulate GIS strategy. They form a basis also for understanding and
interpreting the chapter’s case studies, evolutionary model, and its practical
recommendations.
CASE STUDY: NORWICH UNION
As seen in Chapter 7, the Norwich Union has implemented major uses of

GIS in flood insurance and innovated with GPS installed in cars to give
special lower rates for auto insurance. Norwich has benefited from spatial
technologies. Thus it would be expected that the company and its top man-
agement would recognize GIS and position the firm strategically to take ad-
vantage of it to a greater extent. This continuation of the case study addresses
Figure 12.3 Norwich’s GIS-Based Flood Map for the Flood Plain in Eastern
Shrewsbury, England. Source: Pt.Exsa, 2007
why GIS, although recognized as yielding competitive benefits, has not been
incorporated into Norwich’s strategic plan. It draws on alignment theory and
on parts of situational analysis.
Norwich leads the UK in flood maps and in turn its flood insurance leads the
country. The flood map’s competitive position is underscored by the demand
from competing insurers, reinsurers, and the property market to purchase it
(see Figure 12.3). Norwich has given a “nuanced” response to this demand. It
will sell some of its flood data, but not all. Because Norwich sells three of its
seven flood coverages, the customer is able to discriminate flood risk at only
three levels, versus the ten levels the firm has available internally. In this way,
Norwich has capitalized on a lead product, while retaining some competitive
advantage. Its strong position in flood data also stems from the high prices
and bureaucracy in obtaining government flood data in the UK.
CASE STUDY: NORWICH UNION 363
Norwich’s GPS-based “Pay as You Drive” product (see Chapter 7) is not

only a leader competitively and saves money for customers, but the firm has
also gained a favorable public image, since the new product and its traffic
database improve driving and public safety, while relieving congestion.
Norwich is competitive and profitable from its GIS and spatial products,
and respected for them by the UK insurance industry and government. Thus,
it is surprising that GIS is not included in the company’s strategic plan. The
problem is in large part due to an organizational barrier between the GIS and
IT departments. Norwich’s IT department is responsible for huge databases
located on mainframes, many of them aging systems, and it processes massive
amounts of claims while at the same time it is growing capacity to handle
new business. It has a problem in integrating the massive data across multiple
storage “silos” which restrains company-wide access to some data. This
narrow and legacy-based IT focus has lowered the IT department’s interest in
GIS and diverted attention away from it.
The GIS department has its own strategic plan, but the obstacle is to
obtain the requisite money and support. It depends on IT for storage of
its data and maintenance of its GIS software. However, the problems in
communications between IT and GIS have hobbled the GIS department with
insufficient computer support and resources. It cannot plan big for the future,
even though it perceives the benefits of spatially-enabling extensive data and
developing new applications. Beyond IT, the upper-middle management at the
company has not understood the GIS benefits. As one GIS employee stated,
[Norwich’s upper middle managers] “don’t get it, they don’t understand, and
they don’t have any vision. The are not educated about spatial.” Curiously,
some of the company’s highest executives have visited the GIS department
and are aware of its activities.
Norwich’s situation can be explained using the IT strategic alignment
model. The model advocates corporate and IT strategy alignment (Figure
12.1) and was extended to corporate and GIS alignment. Norwich’s problem
is that, although IT is aligned to a part of corporate strategy and GIS is
aligned to a different part, IT and GIS are dysfunctional with each other.
Since the scale of the GIS systems is much smaller than for IT and because
of the attitudes mentioned, GIS is being overshadowed by IT. Hence, GIS
is viewed from above as part of the overall IT strategy which is somewhat
misaligned with corporate strategy. Although GIS’s lead applications are
competitive and well-known externally, it has low internal visibility and thus
it is not surprising that GIS has not made it into the strategic plan or vision of
Norwich.
What can be done to remedy this situation? A strong-willed top corporate
management could (a) force integration of IT’s data resources into a cor-
porate data warehouse, (b) require spatial enablement of many parts of this
data warehouse, and (c) task the GIS department to provide beneficial and
innovative applications based on the data warehouse, which will be included
in the corporate strategic plan. It would be critical as part of this process for
IT and GIS to start working together.
AN EVOLUTIONARY FRAMEWORK FOR STRATEGIC GIS
This part of the chapter presents an evolutionary framework for GIS Strategy
that takes into account three key dimensions: (1) extent that spatial applica-
tions are customer-facing, (2) extent that geography is part of the industry
or business, and (3) extent that the industry or business utilizes a spatially-
enabled enterprise-wide integration platform.
The evolutionary model was developed by consultation with industry ex-
perts, scrutiny of industry literature (Harder, 1997; Tomlinson, 2003; Barnes,
2005; Guerrero, 2005; Maguire, 2005; Reed, 2005; Sonnen and Morris, 2005),
GIS academic sources (Grimshaw, 2000; Francica, 2005; Longley and Clarke,
2005; Lopez, 2005), and IT strategic literature (Robson, 1994; Huxhold and
Levinsohn, 1995; Papp and Luftman, 1995; Papp, 2001; Hagel and Brown,
2001; Porter, 2001; Brown and Hagel, 2003; Galliers and Leidner, 2003;
Cegielski et al., 2005; Lai and Wong, 2005; Applegate et al., 2007).
The first dimension of the framework consists of the extent that spatial
applications in the industry or company are directed towards a user base that
is predominantly customers (i.e. customer-facing) versus spatial applications
that are directed towards internal users. Internal users include executives,
managers, marketing specialists, middle-level analysts, operations personnel,
sales force, and field workers.
The second dimension, extent of geography as part of the business, refers
to whether the major business products and processes relate closely to geog-
raphy. An example of an industry linked to geography is transportation, for
which the key function of moving goods, inventory, and people is inherently
tied to geography; another example is the utility industry, for which the
products of energy, purified water, and essential materials are provided
through geographic networks of transmission lines, pipelines, and specialized
transport vehicles. Real estate, another obvious example, has land as its cen-
tral element. On the other hand, the legal services industry has slight linkage
with geography for its essential products and processes. In semiconductors,
the vital products and services are usually weakly linked to geography. In
between in extent of geography are industries such as health care, banking,
metals manufacturing, and pharmaceuticals.
The extent that an industry or business utilizes a spatially-enabled web
integration platform refers to whether it is based on “traditional” desktop
or client-server spatial applications versus those based on the web-based
enterprise architecture consisting of web servers, content servers, the internet,
thin and thick clients (Guerrero, 2005; Lopez, 2005; Sonnen and Morris,
2005). The architecture was discussed in Chapter 5 and is shown in Figure
5.1(d). Sometimes this web architecture is extended to include GPS-enabled
AN EVOLUTIONARY FRAMEWORK FOR STRATEGIC GIS 365
Figure 12.4 Evolutionary Framework of Industry Categories by Two Spatial

Dimensions, 1995
portable mobile devices connected through a wireless network. That mobile-

web architecture was also examined in Chapter 5 and is shown in Figure 5.1(e).
In between are mixed architectures, in which industries or companies depend
on a mixture of the traditional technical GIS platform and the emerging web-
and location-based integration platform.
Changes in the evolutionary framework over time are shown circa 1995 in
Figure 12.4, and in 2006 in Figure 12.5. The 1995 arrangement of industries
was based on industry books written in the mid 1990s (Longley and Clarke,
2005; Harder, 1997; Grimshaw, 2000) and by interviews of three GIS industry
experts knowledgeable of the industry in the mid 1990s. The 2006 company
arrangement is drawn from the results of the book’s twenty research case
studies. Figure 12.4 only shows two dimensions since the spatially-enabled
web integration platform was not present, but only the traditional desktop
or client-server. Geography-oriented industries such as transportation and
utilities are shown with higher values on the x axis while ones with much
less spatial linkage such as steel and private hospitals have low values. On
the y axis, industries that are not customer-facing in spatial technologies
include pest management and oil and gas, while industries such as real estate
were moderately customer-facing in GIS, but no industries in the mid 1990s
are highly spatially customer-facing. The deficit of high customer-facing
is because map delivery was still cumbersome, bandwidths were not high
enough for effective interactive mapping, and businesses mostly had not yet
recognized the benefits of providing maps and spatial analysis to customers.
Motion-Based
Technologies
Prudential
Preferred Realty
Extent Spatial is Customer-Facing

Lamar
Advertising
Sperry Van
URS Ness
Rand McNally
Sears
MapGistics Kaiser
Arizona Republic
Southern
Norwich Union
Company
Bay State Engineering Systems
Chico’s
Health Lg Comm Bank
Lg Credit Bank Western Global Integ. Oil
rm d ,
fo le
Exterminator Lg Insur Co.

at ab
Pl n
n -E
io ly
at ial
gr at
Extent of Geography in the Business

te p
In f S
eb t o
W ten
Legend: Levels of spatially enabled web integration platform (Z axis)

Ex
high
medium
low
Figure 12.5 Evolutionary Framework of Firm. Source: Cases, 2006
In 2006, the complexion of industry distributions changed. The web inte-

gration platform appeared, and is added to the framework as the z dimension.
Compared to 1995, there are only moderate changes in some industries with
respect to extent of geography in the business. These changes arose because
some industries recognized and utilized geographic phenomena more. An
example is health care. Many health care providers expanded into networks
encompassing larger geographies. On the y-axis of extent GIS is customer-
facing, there were substantial changes in some industries due to fast, web-
based delivery of enhanced map products in real-time. An example is real
estate, where increasing use of multimedia and map enhancements occurred.
At the low end of the customer-facing scale, some industries remain at 1995
level such as pest management and oil and gas, for which consumers are
not demanding GIS services and the firms are not emphasizing to provide
them. Other industries such as real estate and advertising have become more
customer-facing in certain spatial functions such as automated vehicle navi-
gation (AVN) and GIS in marketing.
In the z-axis dimension of extent spatially-enabled web integration plat-
form, industries such as oil, newspapers, and real estate lead the way, with
continual, powerful, and appealing delivery of maps and spatial analysis to
users and customers. Industries that have mostly not adopted the web-based
CASE FINDINGS FOR HIGH STRATEGIC FIRMS 367
platforms include banks, health care, and insurance. These regulated indus-
tries did see sufficient benefits in spatial applications to justify the high costs
of upgrading from desktop/client-server to web-based enterprise platforms.
They are also held back somewhat by legacy mainframe systems which are
more challenging to connect to the web-enterprise platform. For the research
cases, such obstacles occurred with the Large Credit Bank and somewhat
with Norwich Union.
This evolutionary framework can be expected to change further, since
spatial technologies are rapidly advancing in business due to reduced costs and
realization of the competitive importance of the benefits (Sonnen and Morris,
2005; Francica, 2005). The framework can be useful to managers who conduct
mid- to long-range planning of their spatial architecture and applications.
FINDINGS OF CASE STUDY ANALYSIS
This section looks at the book’s full set of twenty research case studies to
examine how strategic each one is in GIS and why. For many of the firms, it
discusses whether GIS strategy is aligned with corporate strategy. The cases
are grouped by strategic level of GIS and spatial technologies.
CASE FINDINGS FOR HIGH STRATEGIC FIRMS
Global Integrated Oil

For GIO, the global integrated oil giant discussed in Chapter 1, spatial tech-
nologies are applied enterprise-wide, including in exploration, transportation
and storage, refining, environment, marketing, supply chain, and strategic
planning. The company is pushing the technology limits in areas such as
sub-surface mapping, web-based applications, 3-D maps, data warehousing,
and advanced analytic tools. Overall, the sum of all these applications, some
of which have common data-bases, results in highly-strategic GIS. The GIS
manager pointed out that GIS is an important tool in making long-term strate-
gic decisions lease/sale decisions and in realizing the competitive advantage
of geophysical data. Spatial data are used at the top executive strategy level.
GIS and IS work together organizationally and are technically integrated.
Thus GIS and IS strategies are aligned and in synchrony with corporate
strategy.
MotionBased Technologies
This very small Bay-area firm, which was purchased by a middle-sized tech-
nology company, serves performance athletes worldwide, by receiving and
analyzing athletic performance training data uploaded from the athletes’
GPS-enabled devices. The data are input into a database server and are
combined with topographic and routing data from commercial web services.
The MotionBased proprietary spatial analysis software computes customer
profiles which allow these performance athletes to evaluate and compare
their route performance. MotionBased programmers wrote the web applica-
tions and configured the customer-facing web platform. Thus these spatial
technologies constitute the company’s product and are inherently competitive
and strategic. The general manager pointed out that it is the combination
of internet, aggregated data, and spatial technologies that creates competi-
tive advantage. The key is the web services architecture. Also, GIS, IS, and
company strategies are inherently aligned.
Rand McNally
At this large private firm, the world leader in map publishing, GIS is highly
strategic and is discussed as a full case study later in the chapter.
Sears Roebuck
The giant retailer covered in Chapter 7 has extensive spatial technologies
in six areas: routing and deploying service technicians, delivery, warehouse
optimization, marketing, Sears Smart Toolbox automated vehicle navigation,
and capacity management of workforce in service territories. Significant gains
in productivity of delivery and service maintenance have enhanced the com-
pany’s bottom line. As the leading competitor in spatial technologies in its
huge markets, there is strong competitive advantage. The GIS manager in
routing considers that Sears is stronger competitively in spatial technologies
than any competitors, including GE and many smaller-sized ones. GIS is in-
cluded in Sears’ written strategic plan and there is also a GIS strategic plan.
The competitiveness is achieved through greater productivity, lowered cost
for service calls, and reduced staffing costs for a large delivery workforce.
Southern Company
This large parent utility company dominates electrical production and distri-
bution in many parts of the southeastern U.S. As seen in Chapter 9, it moved
from distributed management of spatial technologies that differ among the
operating companies, to an integrated and centralized enterprise-wide GIS,
encompassing all its companies. At the level of the parent holding company,
this led to closer GIS alignment with IT and corporate strategies. The GIS is
internally directed and is planned to maximize enterprise returns on invest-
ment. There is total dependence on GIS for asset management Southern-wide.
Southern’s electrical production is regulated, but the transmission side of the
business is increasingly competitive and open. The strategic approach is to
CASE FINDINGS FOR HIGH STRATEGIC FIRMS 369
partner with leading generic and utility-specific GIS vendors and to seek
enterprise deployment of the technology.
The former head of enterprise GIS points out that strategic benefits are
obtained by lowered costs and improved customer satisfaction. GIS improves
strategic competitiveness in the transmission part of Southern, since it’s un-
regulated, but not in the regulated production part of the company. The GIS
enterprise initiative had executive sponsorship indicating top management
support, even though it’s not yet included in Southern’s corporate plan. As
part of the enterprise project, strong alignment of GIS and IT was emphasized
and achieved.
Arizona Republic
The mid-sized newspaper focuses its spatial technologies on targeted ad-
vertising, as seen in Chapter 2. To accomplish this, it has spatially-enabled
its mailing capabilities to allow for refined addressing as well as market
saturation. The platform combines technical desktop GIS to support object-
oriented programming and web development in-house, with an enterprise,
web-services platform for inside and outside users. The newspaper’s GIS
has become an enterprise, strategic system, although the functional areas of
application are limited. The GIS competitive advantage stems from being
able to pick geographical areas that are suited to a particular advertiser. This
competitive element is included in the newspaper’s corporate strategic plan.
Many other aspects of GIS are more operational i.e. GIS can help in getting
the product out.
Chico’s
This rapidly growing women’s apparel chain has utilized GIS primarily in
real estate location, customer relationship management (CRM), direct mail-
ing, and business intelligence (see Chapter 4). For site location, GIS can
predict the extent of cannibalization of one store from another, leading to ad-
justments in store-performance targets. GIS tracks the sequences and patterns
of shopping locations for high-end customers, personalizing service. GIS is
written into the firm’s strategic plan and aligned with corporate strategy which
is based on customer service and loyalty, and seeking to create new markets.
Competitive advantage has been achieved by better understanding store op-
erations, better planning for new stores, and the servicing better the existing
customer base through CRM methods. There is good alignment and collabo-
ration between GIS and IT, which relates in part to the strong customer-centric
mission.
Lamar Advertising Company

Lamar’s biggest spatial application is to supply maps to its 900-person sales
force that provides them in turn to customers to enable them to decide among
alternative outdoor billboard placements. Other applications are routing of
its maintenance workforce and support for real estate acquisitions. Lamar
utilizes an integrated, web-based platform that accesses mostly commercial
data, albeit with limited spatial-analysis functionality. Since Lamar is unique
among its competitors in having a spatially-enabled national billboard base,
it draws strategic advantage especially in selling. GIS and IT are combined
functionally and organizationally. The corporate and GIS/IT strategies are
aligned, and GIS is included in Lamar’s corporate strategic plan. There is
focus on a long-range vision for GIS and the GIS and marketing people are
working together to consider new spatially driven technologies for billboards.
Prudential Preferred Realty

This Chicago-area residential real-estate brokerage firm applies GIS internally
for demographic analysis and for agent performance profiling by area while
externally customers and consumers have been attracted by the competitive
web portal that allows them to search and view properties with convenient
tools and imagery of the city areas, amenities, and properties. The spatial
applications are entirely based on an enterprise web architecture. Strategically
it allows the firm to “stay ahead of the pack” by creating customer loyalty
through robust, repeatable spatial services. The competitive importance of
spatial systems is not formalized in a written plan but the firm by habit
does little written planning. The firm emphasizes in its corporate strategy
having the technologies available that appeal to customers and retain their
loyalty and repeatability. Geospatial services are regarded as an important
current technology that is a complementary technology piece in appealing to
customers. There is alignment of IT and GIS, since Prudential seeks to be
fast-moving and integrated in presenting technologies to its customers.
MapGistics
This small health-care consulting firm provides geographic mapping appli-
cations to several hospitals with a focus on one in southern California (see
Chapter 10). The most important use of spatial analysis is bed management,
with lesser use in mapping emergency response and a prototype application
of spatial tracking of certain RFID-tagged patients. GIS is inherent in the
firms’ products and is strategic. The ceo pointed out that GIS is strategic at
MapGistics and the bed management product is also in the organizational
strategic plans for certain of its customers. The vision for GIS is to emphasize
environmental scanning to locate hospital chains and independents that are
more modern in outlook and receptive to its spatial product.
CASE FINDINGS FOR MEDIUM STRATEGIC FIRMS 371
CASE FINDINGS FOR MEDIUM STRATEGIC FIRMS
Kaiser Permanente
Kaiser Permanente, the national hospital chain centered in California, applies
GIS in web map services, travel time analyses, travel accessibility studies,
site selection, and some marketing and medical uses (see Chapter 10). GIS is
significant to meet regulatory requirements that dictate classification of areas
as in-service versus out-of-service, understand where clients are located,
determine what disease profiles a region has, and spatially analyze Kaiser
health-care resources including physicians, MRI equipment, and clinics. GIS
is considered strategic for predicting member travel times, locational analysis,
and facilities siting, but less so in the other areas. GIS’s strategic role is
dwarfed by some other overriding health-care issues. Although not in Kaiser’s
strategic plan, GIS undergoes periodic reviews by top management that tend
to align it more with corporate goals.
Large Insurance Company

This large casualty and property insurance firm utilizes GIS mostly for un-
derwriting, catastrophe planning and management, and capacity planning, as
examined in Chapter 3. The spatial applications are client-server based, with
active planning to move to web-based. GIS is essential to firm underwriters
throughout the country to assess risk and price policies. GIS has become
crucial for the catastrophe unit to spatially allocate risk and manage disas-
ter response. Corporate IT is largely legacy-based, which has held back its
alignment with GIS. The executive level is aware of GIS and includes it in
the strategic plan, but it has not seen it as highly strategic and competitive.
URS
URS, the very large integrated engineering design services company with a
global reach, depends on client-server based GIS internally rather than spatial
web services. However, it does develop spatial web services for some of its
clients. GIS and spatial technologies have been of moderate corporate strategic
importance. They are not included explicitly in URS’s strategic plan, but are
implicit in the categories of overall automation and best practices. In the Dot
Com era, they were explicitly included. URS does appear to be getting “back
on track” in giving more support for development of spatial technologies.
One reason is the rapid growth of GIS. Past vision for spatial technologies
has not always been correct such as mapping of underground utilities, but
other spatial vision items such as compliance and regulatory uses have been
realized.
Sperry Van Ness

Sperry Van Ness, the rapidly growing national commercial real estate bro-
kerage firm discussed in the first chapter, utilizes GIS for sales presentations,
property websites, brochures, and internal searching and viewing of proper-
ties. It is entirely based on an enterprise web-service platform. The firm’s
leadership emphasizes the importance of the sales advisors, and generally
plans everything in the short- and mid-term, rather than long-term strate-
gically. To the extent GIS helps the sales advisors be more productive, it’s
important. GIS is of moderate strategic significance and not currently in the
vision of the firm.
Western Exterminator
This middle-sized private firm provides pest control management to residen-
tial and business customers in California, Nevada, and Arizona. The leading
use of GIS is to map the complete customer base for use by the sales force.
GIS is applied to re-align routing and to determine locations for new service
centers. The firm utilizes predominantly commercial web services. There are
only a handful of internal users for the GIS. The moderate-level strategic
advantages of GIS are to target homeowners, locate service centers, and rec-
ognize underserved areas. GIS is not included in the firm’s corporate strategic
plan. However, it is aligned with the corporate strategy that emphasizes mar-
keting and customer service.
CASE FINDINGS FOR LOW STRATEGIC FIRMS
Large Personal/Corporate Bank

As seen in Chapter 3, the case study of this large international bank concerns
its U.S. division. GIS is used for siting and relocation for its rapidly grow-
ing network, trade area analysis, market share study, and ad-hoc thematic
mapping. The GIS group is very small and utilizes a traditional, client-server
set-up. GIS is not regarded as strategic. It’s not in the bank’s strategic plan
and the executives are not focused on it. The small GIS group is aware of
its potential but has not yet been able to convince bank strategists of its im-
portance. GIS is regarded as creating competitive advantage by providing
branch-related customer data to decision makers in an understandable way
and to support service of current customers and attact new ones.
Large Credit Bank

This very large bank, as seen in Chapter 10, has had a monoline product line
in the credit card area. Several years ago, it acquired several medium-sized
regional consumer banks. Although GIS had been only tangential previously,
CASE FINDINGS FOR LOW STRATEGIC FIRMS 373
the acquired banks raised the need to adopt spatial technologies. The initial
spatial applications are site location, real estate evaluation, and marketing for
the acquisitions, plus planning of the corporate campus. Since the acquired
consumer banks are in competitive markets, the siting of their existing branch
locations needs to be evaluated, new sites explored, and geodemographics
performed. The large credit bank, which has been resistant to GIS and held
back by its legacy IT systems, is beginning to apply it in the acquired banks,
but continues to give GIS little recognition so it’s low in strategic level. GIS,
IS, and corporate goals are not aligned.
Norwich Union
Norwich, one of the largest general insurance firms in the UK, was discussed
in Chapter 7. GIS is low in strategic importance as discussed as a full case
study at the beginning of this chapter.
Baystate Health
Baystate Health is a large New England health-care system with three member
hospitals in western Massachusetts (see Chapter 2). Its eight-year-old GIS
program provides GIS for medical and health applications, environment,
spatial statistics, spatial epidemiology, hospital facilities, routing, marketing,
and emergency and disaster planning and response. Since its big applications
were implemented recently, GIS has not yet attained a strategic level in the
organization. It is not in the firm’s strategic plan, nor does it have its own plan.
Baystate as a whole does not yet have a long-range vision for spatial. The
small GIS department has a long-range vision for GIS, but hasn’t yet been
able to convince others to do so. This may change, however, since the GIS
director and champion of GIS, who is the chair of the surgery department,
has his own vision for GIS to become enterprise-wide and is communicating
that to Baystate’s vice presidents,who are attuned to it.
Engineering Systems
ES is a small consulting firm that provides GIS services to local governments
as well as to utility and transport firms. Its in-house GIS is traditional client-
server. Its own systems serve only in-house users, although systems ES has
installed in client firms serve varied external users. Some employees are
assigned to client sites for extended periods, a form of outsourcing. The firm’s
GIS systems products are very tied to geography. Its in-house GIS systems
are not strategic, relative to its consulting competitors, many of which have
more advanced spatial analysis and/or web-based platforms to reach out to
clients. ES’s GIS and IT are aligned, since they are managed together in this
small firm.
SUMMARY OF FINDINGS
The book’s case companies vary on the three dimensions of the evolutionary
framework as shown in Figure 12.5. The results are categorized in Table 12.1.
Size of firm has no consistent association with strategic level in the frame-
work. Some small firms such as MapGistics and MotionBased Technologies
have centered their business on spatial products and services from their found-
ing, so spatial is immediately strategic. Larger firms are older; many were
founded decades before GIS was invented. Their adoption of spatial tech-
nologies has progressed more slowly over time through stages of growth, as
outlined by Nolan’s stage theory in Chapter 3. Medium-sized firms are in
between these, and tended to have adopted and elevated the importance of
GIS and spatial in shorter time spans than the large firms, for instance Chico’s
and Lamar Advertising.
Strategic level is not related to extent that GIS is customer-facing. This
may reflect that many GIS and strategic spatial applications for the case firms
tend to be more intensive internally than externally. Examples of internally-
focused cases are Global Integrated Oil, and Rand McNally. However, extent
that GIS is customer-facing does relate to type of industry. Consumer ser-
vices, billboard advertising, newspapers, and retail are industries that often
serve their customers directly with GIS. By contrast, the oil and utilities
industries are more proprietary about retaining geographic information for
inside use. Furthermore, post 9/11, those industries have been restricted by
government policy from making most types of spatial information publicly
available, such as detailed maps of their network configurations and loads.
The giant Credit Bank is in the very early process of shifting from centralized
service of customers with credit needs that are not spatially-based to serv-
ing emerging retail customers who often need spatial displays, but overall
the bank maintains traditional, legacy habits that are not spatially oriented.
Regarding Western Exterminator, the pest industry has not yet offered spatial
information to its customers who don’t seem ready to accept it.
Spatial strategic level is positively related to extent of geography in the
business. Seven out of ten of highly strategic firms have a high extent of ge-
ography in the business, while those with low strategic levels are all medium
to low in extent of geography in the business. The medium-strategy firms
are in between. Extent of geography in the business is also keyed to industry
characteristics. For the twenty research cases, those with the largest geograph-
ical component come from oil and gas, utilities, real estate, insurance, retail,
and consumer services, while geography is less important for banking and
consulting.
In comparing the extent the twenty firms are highly strategic in GIS to
the adoption of spatially-enabled enterprise-wide web integration platforms,
the two attributes are highly associated for this sample. Five of the six firms
that adopted web-based, enterprise-wide platforms have high strategic levels
for spatial technologies, whereas five out of seven firms in the traditional
c12
TABLE 12.1 Strategic Levels and Model Dimensional Attributes for Case Study Sample
Extent Strategic
Customer- Geography Spatial
Name or description Size Facing Platform in Business Level
Global Integrated Oil large low Web-based, Enterprise-wide high high

Motion-Based small high Web-based, Enterprise-wide high high
Technologies
Rand McNally large medium Intermediate. Client-server, Enterprise-wide, moving to web high high
Sears Roebuck large medium Intermediate. Client-server, Enterprise-wide, moving to web high high
October 18, 2007
Southern Company large very low Intermediate. Client-server, Enterprise-wide, moving to web high high
Arizona Republic medium medium Web-based, Enterprise-wide, limited to circulation and medium high
advertising
14:44
Chico’s medium low Intermediate. Is moving towards more web-based. high high
Lamar Advertising Co. medium high Web-based, Enterprise-wide, moving to more functionality med-high high
375
Prudential Preferred medium high Web-based, Enterprise-wide high high
Realty
MapGistics small low Intermediate. Is partly web-based enterprise-wide medium high
Kaiser Permanente large medium Intermediate. Mostly traditional, client-server medium medium
Char Count=
Large Insurance Co. large low Traditional Client-server high medium

URS large medium Traditional, Client-server med-high medium
Sperry Van Ness medium high Web-based, Enterprise-wide high medium
Western Exterminator medium low Intermediate. Commercial web services med-high medium
Large Commercial Bank large low Traditional Client-server medium low
Large Credit Bank large low Traditional Client-server medium low
Norwich Union large high Traditional, Client-server med-high low
Bay State Health medium low Traditional Client-server low low
Engineering Systems small low Traditional Client-sever medium low
client-server mode have low strategic levels for their spatial technologies.
The seven firms in the intermediate level of spatially-enabled enterprise-
wide web integration platform have medium to high strategic levels (three
medium, four high) for spatial. This strong tie confirms a premise of the book
that the spatially-enabled, enterprise-wide web integration platform is the
direction that spatial technologies are moving to achieve competitive results.
This finding is explained by Porter’s theory that justifies the advantages of the
internet to successful corporate strategy (Porter, 2001). For companies that
are moving in their business overall corporate strategy towards the internet
and e-business, for instance Prudential Preferred Realty and MotionBased
Technologies, IT strategic alignment theory reinforces that the spatial-web-
integration platform corresponds to a corporate strategy of e-business (Porter,
2001; Hagel and Brown, 2001).
Spatial technologies have grown in the business world. Uses are evolving,
supported by newer technology platforms, in particular the spatially-enabled,
enterprise-wide web integration platform. As costs decreased, more profitable
uses were discovered. The extent of customer-facing spatial applications has
steadily expanded over time. This is due both to the convenience and user-
friendliness of the new technologies as well as to broadened user bases. Porter
(2001) concluded that the web services platform encourages greater direct
contact, without intermediaries, between the customer and the product or ser-
vice. This trend is also consistent with findings that web-services platforms are
beneficial (Hagel and Brown, 2001). Porter’s reasons for the competitiveness
of the web-services platform, namely reducing costs, increasing openness,
and lowering barriers to entry, all apply to the spatial realm. Consistent with
his reasoning, GIS can emerge from its departmental niche and serve broad
internal users and/or outside customers. GIS can be included as part of a
broader IT strategy of implementing the web services platform, as long as the
IT and GIS capabilities are present to support it (Henderson and Venkatraman,
1992; Papp, 2001; Cegielski et al., 2005; Applegate et al., 2007).
The extent GIS is customer-facing is not associated with the level of GIS
strategy. This result reflects the differences on GIS’s internal or external
impacts, across a range of firms and industries. For some, such as GIO, the
GIS impacts are highly significant and realized by internal efficiencies and
productivity. At the other end, Sperry Van Ness’s successful GIS strategy
principally benefits the customer. The bottom line is that spatially strategic
firms can be at either extreme or in between.
The extent of geography in businesses overall has grown slowly but
steadily, as the structure of industries altered or new industries emerged.
An example is the web-portal industry sector, which emerged in the last ten
years and in 2005 took a giant step towards the wide public consumption
of GIS with such geographic-based offerings as Google Earth, Yahoo Maps,
and Microsoft Virtual Earth. Porter (2001) observed that a firm with a web
services platform would tend to have a broader reach of geography. For estab-
lished industries such as banking, the trend towards mergers and acquisitions
PRACTICAL ADVICE TO MANAGERS ON STRATEGIES FOR GIS AND SPATIAL TECHNOLOGIES 377
can broaden the geography of the firm and make spatial applications more
important for it. Drawing on the IT alignment literature, the business strategy
for some companies might be naturally keyed to geography, for example Rand
McNally’s business is to produce maps and for Sperry Van Ness’s to map and
visualize commercial properties. Alignment theory would thus point to the
benefit for a corporate strategy tied to geography to be aligned with a GIS
strategy (Papp, 2001; Cegielski, 2005; Applegate et al., 2007).
PRACTICAL ADVICE TO MANAGERS ON STRATEGIES FOR GIS

AND SPATIAL TECHNOLOGIES
The implication from the book’s research case studies is that a company should
evaluate its industry to determine how suitable it is to spatial technologies,
as well as to gauge how much competitive advantage these technologies
offer the firms in the industry (Hagel and Brown, 2001). A firm should
consider deploying spatial applications on web-services platforms, as long as
it can rationalize the investment from a cost-benefit standpoint and support it
technologically. A manager should assess how naturally geographical his/her
business is, to help in determining the strategic potential of GIS for the
company. Once underway with GIS, corporate management should consider
evolving the applications to an enterprise-wide web-based platform. Some
firms such as MotionBased have been able to leap-frog directly to full web-
based, enterprise spatial platforms and take the lead in market niches. Of
course, the springboard approach increases risk and requires a robust business
plan, technological capacity, and understanding of customer markets.
Matching the GIS Strategic Plan with the Corporate Strategic Plan
and the Corporate IT Plan
Another practical implication of the cases for managers is that the alignment of
GIS strategy with corporate strategy is recommended, but with a difference.
GIS applications need to be aligned with corporate strategy and also with
IT strategy. For a company in which IT and GIS are combined together
organizationally such as Sperry Van Ness, IT and GIS strategies can easily
be integrated. However, IT and GIS are organizationally combined for only a
quarter of the book’s cases, which tend to be smaller ones.
For most firms, GIS and IT are separated, so coordination of their strategic
planning might be problematic. Norwich best represents this problem. It has
had limited coordination and communication between GIS and IT. On the
other hand, for firms such as Sears and Rand McNally, GIS and IT work
together well including in coordinating strategies.
The alignment of strategies for GIS, IT, and the corporation is a challenge
that GIS managers must address. Based on the experiences in the book’s case
TABLE 12.2 Practical Steps for GIS Managers to Better Align GIS Strategy
1. Meet and communicate with managers who are responsible for the corporate strategic
plan.
2. If GIS and IT are separate, maintain regular communications with the leadership of the
IT department
3. Consider the extent the company has already emphasized e-business. If the extent is
high, consider incorporating web-based enterprise GIS in the GIS strategic plan if not
present already.
4. Coordinate for future planning of web-based enterprise GIS with the corporate strategic
plan and the IT strategic plan. Coordinate the web and internet emphasis with the other
plans.
5. Consider the infrastructure of GIS, IT, and the business. Are the infrastructures in
sufficient synchrony so all three aspects work together.
studies, steps are recommended for the manager or executive in charge of

GIS, as seen in Table 12.2.
CASE STUDY: RAND MCNALLY
Rand McNally is the famous map company founded in 1856 that has a pre-
mier brand name. It produces maps, atlases, and electronic map products for
consumer, business, transportation, and education markets. Its products are
available in 55,000 stores worldwide. GIS is used integrally throughout the
company to create and update its map and electronic products, support direct
store delivery (DSD), i.e. locating and stocking retail outlines that carry its
products, manage inventory and supply chain, provide specialized maps to its
marketing and planning departments, sell exact maps to the trucking industry,
and support mapping for cell phones and handheld GPS devices. Most of the
map users are internal, centering on the Geographic Information Services De-
partment, but also include people in marketing, sales, supply chain, and other
areas. GIS is recognized throughout the company workforce, from operations
people up to executives. GIS is integral to the firm’s competitive strategy.
Some users are external, including data providers and licensers, supplies,
industry users of electronic products, users of the company’s web services,
consumers who utilize cell-phone map products, and customers of the Rand
McNally GPS Navigator. The firm is the nation’s dominant supplier of truck
routing maps, so much so that Rand McNally is the de facto standard in
that aspect of trucking. Intelli Route Online, shown in Figure 12.6, offers
truckers a web service that provides routes, maps, cost estimating, and HazMat
information.
Rand McNally has integrated its data into large data warehouses accessible
through data mining. The data extracted can be entered into GISs for conver-
sion into maps or electronic products. The data warehouse approach allows
massive amounts of data to be stored and re-used for different applications.
CASE STUDY: RAND MCNALLY 379
Figure 12.6 Rand McNally’s Intelli Route Online showing Phoneix to Denver
Route. Source: Rand McNally, 2007
One of the firm’s biggest GIS functions is in direct store delivery (DSD).
DSD predictive models combined with GIS are able to estimate stocking
requirements for 55,000 retail outlets and deliver target amounts of maps on
a daily basis. It even can estimate by Plan-o-grams the shelf-space layouts
and map placements in the retail stores.
The company is involved in leading mobile and web services applications.
It provides small-sized maps for mobile phones, but not for PDAs. As seen in
Figure 12.7, Rand McNally Traffic transmits metropolitan traffic conditions
and accident reports to cell phone users. Its web services appear as products
on its highly rated website, RandMcNally.com, including a wide variety of
maps and driving directions. Customers can enter online, customized requests
for high-quality, large maps that are mailed.
There is excellent cooperation between the GIS and IS groups. The com-
pany doesn’t underestimate that GIS is very challenging to implement and
“not for the faint-hearted,” so it invests in people and resources for GIS. It per-
ceives that although the web simplifies access to maps, the user tools remain
often challenging, and the often complex outputs may need to be explained
to even sophisticated users.
GIS is regarded as highly strategic and competitive. For instance, in its
central function of producing a map, largely automated tools allow the selec-
tions of a medley of the best of the data from multiple data sources. There
are huge efficiencies in being able to access data warehouses again and again.
The same data can be re-used extensively for a broad range of products from
large, custom-finished wall maps to 1.5 inch cell phone displays.
(a)
(b)
Figure 12.7 Rand McNally Traffic for Mobile Phones. Source: Rand McNally,
2007
Both the company and the GIS department have instituted strategic plans.
They are closely aligned in contrast to the Norwich case. For emerging spatial
technologies, Rand McNally emphasizes being out in front of the vendors,
sometimes building applications in-house ahead of major GIS software ven-
dors such as ESRI. It has focused on improvements in the “basics,” i.e. on
high-quality data from the best data sources that is processed and delivered by
paper and the latest electronics, phones, web services, and other technologies.
CHAPTER SUMMARY 381
The firm has reached a mature integration stage for GIS. It is not surprising
that the company dominates its market niche. GIS contributes by enabling
best practices for producing paper maps, electronic maps and services, and by
attaining efficiency and precision in store deliveries. There is close alignment
between GIS, IS, and corporate strategies. The firm emphasizes emerging
technologies and R&D, and demonstrates vision in GIS.
THE FUTURE AND SPATIAL STRATEGIES IN ENTERPRISES
As seen throughout the book, implementing spatial solutions has required risk
taking, modifying business processes, overcoming resistance, and sometimes
organizational change. At Lamar Advertising, the sales manager’s business
processes were changed by being able to quickly show prospective customers
visual representations of advertising alternatives. At Chico’s, new locational
strategies became available for locating stores through visual display of store
cannibalization, traffic patterns, and sites of the competitors. This has stimu-
lated cross-organizational flows of information.
For the future, new and unexpected inroads of GIS and spatial technologies
can be expected for companies and industries already heavily into GIS as well
as ones with little of it currently. By taking risks and applying spatial appli-
cations to unforeseen business problems, future impacts can be expected on
business processes, employees, and organizations. Several factors discussed
in the book that may influence where GIS and spatial technologies are heading
in the future are shown in Table 12.3.
Because changes in spatial technology are expected, management in orga-
nizations needs to include in strategic planning the assessment of technologi-
cal change. Two-thirds of the GIS managers interviewed from the cases in the
book responded that they did not do this as part of their strategic planning.
In spite of the proximate large time demands of developing and operating the
company’s spatial functions, GIS managers must carve out time and allocate
resources to explore and assess long-range technological change.
CHAPTER SUMMARY
The capacity of a company to implement geobusiness depends on excel-

lence in technology, appropriate spatial applications, an IT base, and skilled
personnel. However, these alone will not lead to durable and lasting success.
Companies having enduring spatial competitive advantage such as Sears made
GIS and spatial technologies a corporate priority. GIS is written into these
companies’ strategic plans. There is knowledge and interest in GIS at the
executive level. Such firms continue to invest in GIS and to research the best
ways to do deploy it in the future.
TABLE 12.3 Trend Factors Likely to Influence Geo-Business in Future Years
Factors in data, architecture, software, technology

r Transition to a more web-centric environment for GIS
r Integration and interoperability of GIS software with enterprise systems. More
enterprise software with built-in spatial functionality
r Geocoding and digitizing of growing amounts of business information
r Expanded public data sets of spatial information worldwide.
r Private, proprietary spatial data sets for a wider reach of business functions, linked to
broader and enterprise-wide attribute data.
r GIS incorporated extensively into mobile devices
r Enhanced enablement of spatial data-bases and data warehouses
r Personal, consumer use of GIS and spatial technologies globally
r A world of increasingly RFID-tagged objects, with greater spatial analysis of
RFID-based data.
r GIS applications available across the entire enterprise in simple formats
r Sophisticated systems to support spatial decision-making and business intelligence
Factors in GIS industry, workforce, management, organizations

r More manufactured products that are GPS-enabled
r Expanded spatial service products, both consulting and web-based.
r Larger and better-trained GIS workforce
r Business education that includes GIS and spatial technologies in standard curricula
r Improved integration even up to merger in organizations of the GIS and IS functions
and departments.
r Expanded selective or complete outsourcing of the GIS and spatial functions
r Stronger security measures in organizations for GIS and spatial applications.
r Broader laws and court precedents that businesses and society can recognize and plan
for.
r Continuing importance of ethical awareness in spatial decision making.
The chapter offers theoretical background, case-study experiences, an evo-

lutionary model, and practical recommendations on how the best strategic
results have been achieved by some, while other firms implement GIS at low
to moderate strategic levels. A key aspect is vision—the leader’s ability to
picture the benefits and advances from spatial systems. The leader cannot
make things happen by himself or herself, but must gain commitment of
others in key roles in the organization.
Theories are helpful in understanding how GIS strategies can be put in
place. Since the book has emphasized the trend towards the internet and
web-services architecture, Michael Porter’s framework helps to explain the
REFERENCES 383
rise of map web services. The open architecture, quick development times,
and flexibility to change rapidly justify this approach.
IT Alignment theory is a mainstay for understanding strategic IS. Since
GIS and IS functions are usually separate organizationally, a variation of
this theory is that GIS planning also needs to be aligned with the corporate
strategic plan and the IS strategic plan.
An evolutionary framework for strategic GIS is given. The dimensions ex-
amined are extent to which GIS is customer-facing, extent of geography in the
business, and extent of a spatially-enabled, enterprise-wide web intergration
platform. The latter two factors are associated positively with a high level of
strategic GIS. Another factor not associated with strategic level is size of firm.
New, small companies such as MotionBased can jumpstart GIS and spatial
technologies as their core strategy, while mature ones such as Sears can evolve
more slowly to achieve more importance for GIS and spatial technologics.
Practical steps are recommended for leaders and managers of firms seeking
a strategic approach to GIS. Will and tenacity are necessary to raise a firm’s
strategic competitiveness in spatial technologies, as they are new and may be
disruptive to traditional approaches. Assessing future spatial technological
change needs to be a part of the GIS manager’s agenda.
Geo-Business: GIS in the Digital Firm has approached GIS from the per-
spective of business and industry. While government has historically been the
major economic sector for spatial technologies, GIS in the private sector is
on the ascent. The book has organized its contents according to the context of
the business manager or GIS specialist. Relative to many book approaches,
geography and spatial science have been covered less, and the frameworks
and empirical experiences of businesses with GIS and spatial technologies
emphasized more. A core foundation has been provided in GIS technical
principles, mapping, spatial databases, architecture, and spatial technologies.
Although future changes in the technologies will be fast and likely dramatic
at times, one goal of the book is to cover the “softer” spatial areas as well
of decision-making, marketing, management, systems development, intangi-
ble benefits, organizational change, industry differences, outsourcing, ethics,
law, managing security, and strategy. Compared to technology, the content of
these areas will likely prove more durable over time, but also subject to the
continual need to learn and update. In the end, it is hoped that managers and
practitioners of Geo-Business will understand the knowledge and experiences
of this field, set lofty goals, and move their enterprises forward to provide
benefits to their stakeholders and to society.
REFERENCES
Applegate, L.M. 2007. Corporate Information Strategy and Management, 7th Edition.
Boston: McGraw-Hill.
Barnes, S. 2005. “Beyond the niche.” Geospatial Solutions, May 1.
Brail, R.K., and R.E. Klosterman (Eds.). 2001. Planning Support Systems. Redlands,
CA: ESRI Press.
Cegielski, C.G., B.J. Reithel, and C.M. Rebman. 2005. “Emerging information tech-
nologies: Developing a timely IT strategy.” Communications of the ACM, 48(8),
113–117.
Daratech. 2004. Geographic Information Systems Markets and Opportunities.
Cambridge, MA: Daratech Inc.
Dun and Bradstreet. 2005. Dun and Bradstreet Small Business Solutions. New York:
Dun and Bradstreet.
ESRI Inc. 2004. “Law Enforcement and the GIS Enterprise.” Available at
www.esri.com.
Francica, J.L. 2005. “GIS and the Future in Business IT.” In Pick, J. (Ed.), Geo-
graphic Information Systems in Business, Hershey, PA: Idea Group Publishing,
pp. 358–372.
Galliers, R.D., and D.E. Leidner (Eds.). 2003. Strategic Information Management.
Oxford, England: Butterworth/Heinemann.
Ghosal, S. 1987. “Global Strategy: An Organizing Framework.” Strategic Manage-
ment Journal, 8, 425–440.
Guerrero, I. 2005. “Emerging technologies in the geospatial industry.” In Proceed-
ings of the Annual Conference of the Geospatial Information and Technology
Association, GITA: Aurora, CO.
Hackbarth, G., and B. Mennecke. 2005. “Strategic positioning of location applications
for Geo-Business.” In Pick, J. (Ed.), Geographic Information Systems in Business,
Hagel III, J., and J.S. Brown. 2001. “Your next IT strategy.” Harvard Business Review,
October, 105–113.
Hendriks, Paul H.J. 1998. “Information Strategies for Geographical Information
Systems.” International Journal of Geographical Information Science, 12(6):
621–639.
Hoover’s. 2005. Hoover’s Online. Hoover’s Inc., Austin, TX.
Huxhold, W.E., and A.G. Levinsohn. 1995. Managing Geographic Information Sys-
tem Projects. New York: Oxford University Press.
Jarupathirun, S., and F. Zahedi. 2007. “Exploring the influence of perceptual factors
in the success of web-based spatial DSS.” Decision Support Systems. Article in
press and available in proofs at www.sciencedirect.com.
Lai, V.S., and B.K. Wong. 2005. “Business types, e-strategies, and performance.”
Communications of the ACM, 48(5), 80–85.
Laudon, Kenneth C. 2005. Management Information Systems, 9th Edition. New York:
Prentice Hall.
Longley, P.A., M.F. Goodchild, D.J. Maguire, and D.W. Rhind. 2005. Geographic
Information Systems and Science. New York: John Wiley and Sons.
Lopez, X.R. 2005. “Location-based services.” In Hassan, K., and A. Hammad (Eds.),
Telegeoinformatics: Location-Based Computing and Services, Chapter 6, Boca
Raton, FL: CRC Press.
REFERENCES 385
Luftman, J.N. 2001. “Assessing Business-IT Alignment Maturity.” In Papp, R.

(Ed.), Strategic Information Technology: Opportunities for Competitive Align-
ment, Chapter 1, Hershey, PA: Idea Group Publishing, pp. 105–134.
Maguire, D.J. 2005. “Enterprise geographic information servers: a new information
system architecture.” In Proceedings of the Annual Conference of the Geospatial
Information and Technology Association (GITA), GITA: Aurora, CO.
McFarlan, E.F. 1984. “Information technology changes the way you compete.” Har-
vard Business Review, 62(3), 98–103.
Murphy, L.D. 1996. “Competing in space: the strategic roles of geographic informa-
tion systems.” In Proceedings of the Association for Information Systems, AIS,
Atlanta, GA.
Nolan, Richard, and F. Warren McFarlan. 2005. “Information Technology and the
Board of Directors.” Harvard Business Review, October, 1–10.
Papp, R. 1995. “Determinants of Strategically Aligned Organizations: A Multi-
industry, Multi-perspective Analysis.” Ph.D. Dissertation. Hoboken, NJ: Stevens
Institute of Technology.
Papp, R. and J. Luftman. 1995. “Business and I/T Strategic Alignment: New Perspec-
tives and Assessments,” Proceedings of the Association for Information Systems,
Atlanta, GA: Association for Information Systems. Available at home.aisnet.com.
Papp, R. 2001. “Introduction to Strategic Alignment.” In Papp, R. (Ed.), Strategic
Information Technology: Opportunities for Competitive Alignment, Chapter 1,
Porter, M.E. 2001. “Strategy and the internet.” Harvard Business Review, 79(3),
63–78.
Reed, C. 2005. “The spatial web.” White paper, available at www.opengis.org, Open
Geospatial Consortium, Wayland, MA.
Sonnen, D., and H. Morris. 2005. “ESRI: extending GIS to enterprise applications.”
White paper, February, International Data Corporation, Framingham, MA.
Standard and Poors. 2005. Stock Reports. Short Hills, Standard&Poors, McGraw-
Hill, New York.
Tomlinson, R. 2003. Thinking About GIS: Geographic Information System Planning
for Managers. Redlands, CA: ESRI Press.
Index JWPR043-Geo-Business October 23, 2007 9:31
INDEX
ABC Blinds 174 Banking 311–312

ABC Company 134–137 Bay State Health (case study) 40, 51,
Agriculture 314–317 53-54, 176, 317
Agricultural Computer for Field application of GIS 54
Operations and Decision-Making Benefits of GIS 201
315 Benefits of GIS Sample of Firms
Airport Runways with Noise Plumes 206
284 Benefits of Mobil Mapping, Based on
Albert R. Sparaco v. Lawler, Matusky, Lessons Learned at Xcel Energy
Seklly Engineers 343 275
Altura Credit Union – Market Potential BI Example in Insurance with Web focus
and Competitor Analysis 313 from Information Builders 92
American National Standards Institute Billboard Effect 310
(ANSI) 163 Broadband Penetration Rates by Country
Analysis 171, 176–178 21
Application Program Interfaces (APIs) 23, Buffer Analysis 61
64, 144, 145 Business Continuity Plan 277,
definition of 138 297
Appropriate Data Model for Certain Data Business Fleet Routing and Scheduling
Modeling Situations in Business 239 281
Architecture Design 179 Business Intelligence (BI) 70, 74, 78–79,
Architecture of SDSS for Insurance 252–253
Pricing of Typhoons 87 WebFocus software See WebFocus
Architecture of Seaspan’s Mobile Business Model of Typical Electric or Gas
Network 153 Utility 268
Arizona Republic xv, 369 Business-GIS-IT Strategic Alignment
Auto CAD 127 Model 359
Autodesk Inc. 26 Business-IT Strategic Alignment Model
Automatic Vehicle Location 190, 281 358

Copyright © 2008 John Wiley & Sons, Inc. 387
388 INDEX
Business-to-Business (B-to-B) 3, 134, Comparison of Cost-Benefit for a Simple

156, 159 GIS System to a GIS System Closely
influences from GIS and spatial Coupled with Another System Or
technologies 157 Technology 198
Business-to-Consumer (B-to-C) 3, 4, 134, Comparison of Web Services and
158 Client-Server Models 137
influences from GIS and spatial Computer-Aided Design (CAD) See Auto
technologies 156 CAD
Computer-Assisted Software Engineering
CAD See Computer-Aided Design, Auto (CASE) 172
CAD Computer System to Identify Local
CatNet–Tracking of U.S Hurricane Paths Resources 345
301 Concepts of Longitude and Latitude
Call Before You Dig Management for 46
Water Utility 272 Conceptual Model of SDSS
Cardinalities 180 Task-Technology Fit 82
Cartographic Design 183 Copyright 342
Case Study Firms in GeoBusiness xv, Cost-Benefit Analysis (C-B Analysis)
369 196–214
Case Study Firms Using Formal Systems analysis of feasibility 213
Development 188–189 analysts 209
Case Study Results on GIS Linked to challenges 197
Enterprise Systems 103, 104 comparison over time 210
Catalog Servers 138 context 196
Cell Phone Used for City Navigation and criterion 211–212
Information 162 in case study companies 202–208
Census (See U.S Census) intangible 212
Chico’s (case study) xv, 115, 116–117, method of discounted cash flow
369 211
Chi-Square Automatic Interaction Sears Roebuck See Sears
Detector (CHAID) 303 stakeholders 213–214
China 86–89 substeps for IT 208–210
City of Portland 124 unit of analysis 207
Classifications of a DSS 75 Costs of GIS for Case Sample of Firms
Classification of Spatial and Non-Spatial 204–205
Information Systems 83 Coupling technology 17
Client-Server Architecture 134–138 Crime 347–350
strengths 135 Criteria to Evaluate Spatial Data in
thick client 134 System Design 182
thin client 134 Criterion 211
Client-Server, Spatial Web Services, and Customer-to-Customer (C2C) 156
Mobile Services Architectures Customer Relationship Management
135 (CRM) 5, 20, 42, 65, 99, 110-115,
Cold War 17 303
Community Tapestry Categories,
Arranged by LifeMode Groups DS Waters 149
305 Dangermond, Jack 15
Community Tapestry Income Range for Data (definition of) 43
LifeMode Groups 306 Data Flow Diagram (DFD)
Community Tapestry LifeMode Summary Data Flow Diagram for Marketing Map
Groups by U.S County 307 (fig.) 179
INDEX 389
Data Management Edens & Avant (E&A) 142, 155

introduction to 228–231 Edens & Avant Spatial Web Services:
quality management 256–258 Property in Port Plaza Shopping
Data Mining 123, 253 Center, New York 143
Data Warehouses 122–123 Ebay 38, 156
definition 248 Electric Data Exchange 2
extraction, transformation, and load Electronic Commerce (e-commerce) 3,
(ETL) 123 115, 155–161
fundamentals 249–254 Electronic Product Code (EPC) 19, 20
future directions 255–256 Engineering Systems (case study) xv, 190,
history 249 373
key applications 254–255 Engineering Systems Map of Land Update
metadata 123 and Maintenance for Fairfax County,
physical structure 253 Virginia 191
spatially enabled commercial data Enmax (case study) 227, 246–248
warehouses 253 Enmax Corporation Substation 246
Data Warehouse and Its Data Flows 123, Enmax Database Configuration 247
251 Enterprise Resources Systems (ERP) 5, 8,
Data Warehouse Star Schema, with 30, 65, 99, 208
Location Included 251 alternative means of GIS integration
Dataplace.org 145–147 107
Dataplace.org Map of Vacancy Rate in ERP II 106–107
Evanston, Illinois 147 GIS integration 102–110
Decision Support Systems (DSS) 70–79, motivations for moving to 102–103
83, 90 Enterprise Web Architecture 32, 136, 229,
executive information system 252
74 Enterprise Wide Approach 101
group decision support system Entity-Relationship Diagram (ERD) 179
74 Entity-Relationship Diagram Example 180
history 77 ERP with Basic Features 105
Delaware Department of Transportation ERP with Full Features 106
(DOT) ESRI xiv, 5, 26, 79, 108, 125, 152
Dell Computer 141 ArcFM 108, 127
Deontological Theories 329 ArcGIS 26, 76, 77, 91, 92, 115, 127,
Design 171, 178–184 243, 245, 251, 254
Design for Controls and Security ArcIMS 93, 108, 142
183 ArcInfo 15, 16
Design for the Acquisition of Spatial Data ArcPad 17
181 ArcSDE 108
Design of Data Standards and Data ArcView 3-D 47
Conversion 183 ArcWeb Services 90, 155
Diagram of a DSS 72, 76 ArcWebExplorer 145
Dialectic 329 Business Analyst 26, 142
Digital Photo and Associated Satellite geodatabase (.MDB) 161
Image Displayed Together 60 geodatabases 183, 243–246
Digitizing 55 geography network portal 55, 182
Digitizing Mouse and Tablet 56 history 15
Direct Marketing 124 market control 26
Dot.Com era 3, 31 ModelBuilder 183
Dow Chemical Company v. United States RouteMAP 142
346 shape file (.SHP) 166
390 INDEX
ESRI Inc. Founder and President, Jack Geodata.gov Map Viewer Display of
Dangermond 15 Transportation Network that Includes
Ethical Issues 328–336 Kansas City and Wichita, Kansas
dilemmas and issues 330–331 55
Ethical Relativism 328–329 Geodatabases 299
European Committee for Standardization Geodemographics 302–311
163 application to spatial analysis 306
Event-driven Architecture (EDA) definition 304
23–24 functions 304
Evolution of the IT Context for GIS, privacy controversy 338
1965–2005 Geodetic Function See Spatial Functions
Evolutionary Framework 364–366 Geographic Markup Language (GML) 4,
Evolutionary Framework of Industry 140, 141, 164–165
Categories by Two Spatial Geography Network, The 54, 210
Dimensions 365 Geolocation Technologies 28
Example of Location Query in a Spatial Geologic Model of Oil and Gas Deposits
RDBMS 242 in the Cretaceous Period 10
Example of Relational Tables and Geomedia 254
Sequence of Operations by Three Geospatial One-Stop (www.GeoData.gov)
Relation Operators on Supplier Data 54, 182, 210
231, 234 GeoWeb 64–65
Example of Zillow.com Make Me Move GIO’s Supply Chain Routing of
Page 159 Containers from Plant to Customer in
Exclusive OR 240 Africa, South Asia, and Australia
Extensible Markup Language 22 13
Extranet 136 Graphic Information Systems (GIS)
alignment of strategies 377
Fast Food Sales Regions and analysis 58
Transportation of Network: Example Autodesk 247
of Spatial Analysis 50 benefits 200–202
Factor 357 budget analysis 207
Federal Geographic Data Committee buffering See Buffer Analysis
163 code of ethics 338
File Design to Represent a Theme in costs 199–200
ArcGIS 245 cost-benefit analysis reasons 196
Finance 311–312 data 44
Formal Controls 349 decision making 71
Formal System 348 departments 264
Formal and Informal Security Controls for emerging spatial market 30
GIS and Spatial Technologies 349 free availability of data 210
Forward Looking Infrared Technology future trends 65
(FLIR) 346 GeoLink 108–110
Framework of Enterprise Applications 100 history 14–17, 24, 173
linkages to other disciplines 51
Generic Design of a GIS, Showing model 6
Attribute Data and Digital Boundary organization 44
Data 7 outsourcing See Outsourcing
Geo-referencing 5 people 44
Geo-surveillance 344–345 risk of application 197–198
Geodata Analysis and Display System satellites 58
(GADS) 77 standardization 161–166
INDEX 391
technology 44 Human Resource Management 100

training 44 Hurricane Katrina 64, 93
workforce 262
GIS/Spatial Software Worldwide Imagery Used in GIS 59
Revenues from Leading Vendors 26 Implementation 171, 184–185
GIS Code of Ethics 338–339 Important OGC Standards 164
GIS, DSS, and SDSS 76 Information (definition of) 43
GIS in GIO’s Retail Market Planning in Information Systems (IS) 4, 40–43
Africa 12 data transformation into info.
GIS and Homeland Security: Threats to Knowledge 42–43
Utilities 277 definition 42
GIS and IT Factors in CRM 116 goals 41
GIS and Satellite Overlay of 400 kV non-spatial IS 197
Transmission Line of Great River systems development process 42–43
Energy in Minnesota 270 technology 41
GIS Specialist at Work on a Business Informal Controls 349
Problem 39, 65 Informal System 348
Global Grid Computing Environment Information Systems Model for Business
166 Problem Solving 40
Global Integrated Oil (case study) xv, Information Technologies (IT) 1, 197
9–13, 31, 203, 367 Information Technologies Alignment (IT
Global Integrated Oil’s Energy Value Allignment)
Chain 10 Inheritance 229
Global Mapper 155 Insurance 296–298
Global Positioning Systems (GPS) 6–7, Insurance Risk Levels and Spatial
24, 120, 147, 183, 198, 216, 230 Decision Analysis in Fourteen
enabled cars 220 Regions of Guangdong Province,
history 17 China 89
Go-Geo! 210 Intangible 197, 203
Google Intangible Costs 203, 207, 212–213
Earth 3, 23, 24, 31, 86, 145, 330 Intellectual Property 342
Maps xvi, 141, 142–143, 155, 158 IntelliRoute Online Showing Phoenix to
Gramm-Leach-Bliley Act 350 Denver Route 379
Graphical User Interfaces (GUIs) 15, 16, Intergraph 26–27
24, 72 International Business Machines (IBM)
Grid Computing 165–166 30, 77, 162
International Organization for
Hagel 141 Standardization (ISO) 163
Halamka, John M.D 333 Simple Feature Access 165
Halamka, John M.D of Harvard Medical Internet Strategy 360
School Viewing His RFID Implant in Interoperability 161
2004 334 Intranet 136
Handheld Computer Running ArcPad
Software 18 J2EE Servlet 142
Handheld GPS Unit Showing Redlands Jewish Hospital and St. Mary’s Health
Streets and Business 60 Care Locations 318
Hardened Trimble GPS Device 57 Joint Application Design (JAD) 191,
Health Care 317–319 193
Huff Trade Analysis 288–290
Huff’s Model Application to Estimate the Kaiser Permanente xv, 77, 182, 207,
Trade Area for Micmac Mall 290 321–323, 371
392 INDEX
Kaiser Permanente Estimated Travel Longitudinal Change Detection 62

Times to End Stage Renal Disease Losses and Insurance Claims from
Treatment Sites 322 Hurricane Katrina, by state 301
Kaiser Permanente Forecasted Density of
Membership in Service Areas 234 M2M 112
Kern River Gas Transmission Co. v Costal Maintenance 171, 185
Corp 343 Major Vendors by Category for Mobil
Key data Management Issues for GIS 230 Services 150
Key Data Management Issues for MIS Managing Spatial Data See Data
228 Management
Knowledge Map Layers
Knowledge Engine 72 attributes 48
Knowledge Worker 72, 74 feature ID 48
Kyllo v. United States (case study) Map Overlay Example 61
346–347 Map Overlay Functions 61
Map Overlay of Exploration Field
Law (definition) 339 Topography and Locations of
Lamar Advertising xv, 202, 369 Proposed Pipelines and Well Sites 11
LAN (See Wireless LANs) Map Showing Ship Located in Vancouver
Large Credit Bank (LCB) (case study) xv, Region and Assigned Tugs 154
202, 312–313, 372–373 MapGistics xv, 296, 370
Large Insurance Company (LIC) (case MapInfo 27, 79, 162
study) xv, 65, 90–95, 203, 207, 371 AnySite software 117
Large Personal/Corporate Bank (LPCB) PSYTE 305–306
(case study) xv, 79-81, 114, 262, MapServer 23
372 Market Efficiency 160–161
Leading GIS Applications for Electric and Marketing 302–311
Gas Utility Firms, 2004 273 target marketing 303
Leading Organizations for Spatial Mash-ups 86, 144, 145, 214
Standards 163 MasterCard 134, 157–158
Legacy Systems and Integration 140, 215 Matrix of GIS Applications at GIO by
Legal Aspects of Surveillance: Kyllo v. Business Sector and Life-Cycle State
United States (case study) 346–347 14
Legal Issues 339–346 McFarlan Strategic Grid 31
application of common law to GIS 339 McFarlan Strategic Grid
areas of international law and GIS 341 McHarg, Ian 15
major areas 341 Microsoft 138, 162, 253
Lexus-Nexis 338 MapPoint 33, 144
LIDAR 20, 24 Microsoft Visual Earth 23, 24, 33,
Linux 23 145
Location Based Services (LBS) See Milacron 156
Mobile Services Mobil Devices for Airport Inspection and
Location Privacy Protection Act of 2001 Maintenance 285
336 Mobile Field Display of Neighborhood
Location Quotient 62–63 Utility Network 274
Locational Query for a Spatial DBMS Mobile Services 136, 147–149, 213, 273
242 location-awareness 149
Location-Awareness 149 Model of Geospatial Markets in 2004 30
Logical Components of Routing System Model of the GIS Market 28
for Oversize, Overweight Vehicles in Model of GIS in Precision Agriculture
Delaware 144 315
INDEX 393
Modeling and Forecasting 62 Oracle 30, 112–113, 161–162, 210, 253

MotionBased Technologies xv, 154–155, Oracle Spatial 238–241, 248, 253
173, 367–368 advantages and disadvantages 241
Moore’s Law 41 SQR 241
Moore’s Law. Transistor Capacity of Intel Warehouse Builder 254
Processor Chips, 1971–2006 1.2 Oracle Spatial 10g 239–243
Oracle Spatial 10g – Advantages and
NationalAtlas.gov 210 Disadvantages 244
National Spatial Data Infrastructures Oracle Spatial 10g: Capabilities and
(NDSIs) 64 Functioning of a Spatial RDBMS 240
NavStar 148, 161, 285 Oracle Spatial 10g Integrated Data
NavTeq 155 Architecture for New York City 243
Neighborhood Information Field (NIF) Orchestration of Web Services for Model
309 of Land Susceptibility 140
Nesa (case study) 124–129 Organizational Hierarchy for SDSS and
Nesa’s Corporate IT Application Spatial BI 73
Configuration 125 Organizational Structure 262–264
Nesa’s IT Processes for Outdoor Lighting Outages in Eastern Pennsylvania from
Customer Information Center 127 Hurricane Isabel
Net Present Value (NPV) 195, 203, 207, Outsourcing 28, 41, 178, 193, 203,
215 319–321
Nolan, Richard 85–86, 265 definition 320
Nolan’s Stage Theory – 1979 Version 86 Overall Design of BI from Information
Nolan’s Stage of Growth for GIS/Spatial Builders and ESRI 91
Technologies 266
Non-Spatial IS 209, 212 Parallel Stages in Evolution of GIS and
Normative Value of Information 218 Decision Support Systems 78
Norwich Union (case study) xv, 196, Patent 343–344
220–221, 361–346, 373 Paz Oil 115
Pay as You Drive 220, 363 Pedestrian Navigation 160
Norwich’s GIS-Based Flood Map for the Perceived Value of Information 218
Flood Plan in Eastern England 362 PEMS Data 182
Networked sensors 20 Personal Device Assistants 20
Number of Internet Hosts, 1984–2005 3 Physical Design 181
Pidpa (case study) 107–110
Object 228 Pidpa: User Selection of Calamity Point in
Object-Oriented Systems Development GeoLink Browser Showing Link to
Approach 180–181 SAP Work Orders 111
Object-Oriented Data Model – Pipda’s ERP System 109
Advantages and Disadvantages 238 Pipda’s GIS System Design 110
Object-Oriented Model 228, 229, 235–238 Pipda’s Links of GIS with Major Business
Object-Relational Model 238–239 Systems 112
Online Analytical Processing (OLAP) Pidpa’s Water Distribution Locations in
123, 124, 248, 249, 252–254 Antwerp Province, Belgium 108
Spatial model (SOLAP) 227, 252 Pizza Hut 4
OnStar 148, 285 PlaceBase Inc. 147
Open Geospatial Consortium (OGC) 4, Planning 171, 174–176
138, 140, 163, 165 Point Map 288
influence on spatial web services 164 Points, Lines, and Polygons 48
Open Source 22 Portable wireless devices 20
Open Systems 166 Portland See City of Portland
394 INDEX
Practical Steps for GIS Managers to Better SAP 5, 105–106, 108, 112–113, 115, 125,
Align GIS Strategy 378 253
Privacy 336–339 Sarbanes-Oxley Act of 2002 350
Privacy Protection Act of 1974 336 Scanning 56
Productivity Paradox 196, 218–219 Seaspan Tug Hauling a Barge Offshore
Prototyping 183 Canada 153
Proximity Analysis 61 Security 347–350
Prudential Preferred Realty 370 Seaspan International Ltd. 152–154
Public Safety Answering Point (PSAP) Sears (case study) xv, 196, 203, 214–217,
148 368
Sears Delivery Truck and Driver 282
Radical Analysis 288 Sears Integrated GIS Systems for Product
Rand McNally xv, 102, 356, 368, 378–381 Repair Services and Home Delivery
Rand McNally Traffic for Mobil Phones 215 909 496 3304
380 Sears Smart Toolbox Routing Feature
Rapid Application Development (RAD) 283
172 Semi-Structured Decision 71, 74
Rapid Prototyping 172 Service Oriented Architecture (SOA) 20,
Realistic Value of Information 218 22–23, 93, 106
Referent Disciplines for Concepts and Simple Object Protocol (SOAP) 22
Theories of GIS 52 Situational Analysis 360
Regional Transmission Lines in Smart Trolleys 160
Mid-Atlantic United States 270 Smart Trolley 160
Relational Database Model – Advantages Southern Company xv, 203, 278–280,
and Disadvantages 235 368–369
Relational Model 231 Spatial Analysis 50, 51, 59
difference 232 Spatial Analysis of Industrial Locations
Intersection 232 for Los Angeles Using Location
join operator 231–232 Quotient 63
select operator 231–232 Spatial Business Intelligence (Spatial BI)
union operator 231–232 71, 75, 78, 80
Relational Operators 228 cost 85
Relationship of Spatial Data and Attribute definition of 75
Data 49, 233 examples 90
Requests for Proposals (RFP) 28 telecommunications industry 114
Retail 286–292 WebFocus software See WebFocus
Retail Attributes Data Used for Spatial Spatial Data
Applications in the Retail Industry attribute data 6
287 characteristics 182–183
Retail Study of Loyalty Card Program difference from traditional data 40
309, 310 digital map boundaries 6
Retail Trade Area 288 scope of 8–9
Radio Frequency Identification (RFID) Spatial Data Infrastructures (SDIs) 64,
xvi, 9, 17–20, 24, 28, 31, 102, 65
120–122, 147, 148, 160, 183, 184, Spatial Decision Support Systems (SDSS)
199, 213, 230, 344 70, 75, 78-79, 80, 81-90, 142, 265,
ethical issues 331–335 298
history 17 compared to DSS 90
RFID Tag Close-Up View for Injection cost 85
under Patient’s Skin 332 definition of 75
ROI 173, 203, 207 hazards and insurance phenomena 88
INDEX 395
predicting typhoon insurance risks in Systems Development

China 86 by professionals and scientists 175
task-technology fit 81 case study findings 185–190
visualization 82–84 Phases 171
Spatial Data Warehouses See Data Systems Development Life Cycle
Warehouses (SDLC) 171
Spatial E-Commerce: Ordering a Pizza Systems Integrators 28
Online from Google Map Display 4
Spatial Functions in XML for a Spatial Tangible Costs and Benefits 203, 207, 216
RDBMS 241 Tangible Costs for GIS 200
Spatial Object and Example 236 Target Site for Marco’s Pizza in Las Vegas
Spatial Objects Showing Multiplicity and 288
Class Hierarchy 236 Technical Security Controls 349
Spatial Overlay 14 Technical Security Controls for GIS and
Spatial Relational Database 239 Spatial Technologies 348
Spatial Web Services Architecture See Technical System 348
Enterprise Web Architecture Technologies Often Coupled with GIS 19
Spatial Web Services Provided by Technology Acceptance Model 144
Enterprise-Wide Internet Bus 22 Technology Seminar for GIS Planning 177
Spatial Technologies 319–321 TeleAtlas 158, 185
Spatially-Enabled Enterprise Database Thermal Imaging See Forward Looking
Products 29–30 Infrared Technology
Specialized Hardware Tomlinson, Roger 171
Sperry Van Ness xv, 3–5, 31–34, 64, 114, Tomlinson’s Ten-Stage GIS Planning
159, 207, 360, 372 Process, Compared to Systems
Staged Development 141 Development Phases 172
Stages in Adoption-Diffusion Theory, Total Tile Failures in Unit-11 of Great
Measured by Cumulative Percentage River Energy’s Pleasant Valley 269
Adoption 308 Trade Secret 343
Standardization 161–166 Transportation 280–286, 319
Statistical Analysis 62 Trend Factors Likely to Influence
Strategic 357 Geo-Business in Future Years 382
Strategic Levels and Model Dimensional Trulia.com 145
Attributes for Case Study Sample 375 Trulia Real Estate Search 146
Structural Damage to Superdome in New Turnaround 357
Orleans from Hurricane Katrina 302 Twelve Job Roles of GIS Professionals
Structural Elements of ESRI’s 263
Geodatabases 245 Types of Decision Support Systems
Structured Decision 71, 74, 172 Models 73
Supply Chain Management 5, 20, 118–122 Typhoon Insurance Risks in China 86
factors 119 Tyrwhitt, Jacquiline 14
GIS contribution 120
problems and challenges 120 United States Census
U.S military 121–122 data and mapping 145, 209
Support 357 description 54
Swiss Reinsurance Company (case study) GIS and satellites 7
298–301 TIGER 183
SwissRe’s Available Data in GEOdatabase website portal (www.census.gov) 54,
300 181, 311
SwissRe Headquarters Building in Zurich, United States Center for Geographic
Switzerland 299 Information and Analysis 78
396 INDEX
United States Federal Privacy Legislation WebFocus 91–92

Significant for GIS 337 Web Catalog Service (CAT)
United States Military 164–165
Cold War 15 Web Feature Services 4
CENTCOM 121 Web Map Service 4, 164
DARPA 2 Web Service Architecture See Enterprise
DOD 2, 18, 265 Web Architecture
supply chain 121–122 Web Services 138–142
Unstructured Decision 71, 74 adoption 141
Urban and Regional Information Systems Edens & Avant (E&A) 142
Association (URISA) 338–339 finding desired 138
URS Corporation xv, 177, 184, 203, 371 home utility repair 140
User’s Perspectives on Web Services pluses and minuses 141
139 Western Exterminator xv, 372
Utilities Industry 267–278 Wireless LANs 20, 28
Utilitarianism 329 Wireless Privacy Protection Act of 2003
336
VeriChip Roam Alert System 335 World Wide Web 2
Vision 356 Work Trak 150
Visualization of a Park in London 47
XOR See Exclusive OR
Wal-Mart xvi, 18–20, 122, 216 XML 142, 164
Web-Based Customer Application to
Access Retail Outlet Information 292 Yahoo! Maps 145, 330
Web-Based Enterprise Wide Applications
and Infrastructure 29 Zillow.com 134, 158–159
Web-Based Ring Analysis for Prospective Zillow.com Z-Estimate of Property
Retailers 289 158

Geo-Business GIS in The Dig PDF

Uploaded by

Copyright:

Available Formats

Geo-Business GIS in The Dig PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Geo-Business GIS in The Dig PDF

Uploaded by

Copyright:

Available Formats

fm JWPR043-Geo-Business October 19, 2007 12:4

GEO-BUSINESS: GIS IN THE

Geo-Business GIS in the Digital Organization James B. Pick

This book is dedicated with thanks to my wife

GEO-BUSINESS GIS IN THE

JOHN WILEY & SONS, INC.

This book is printed on acid-free paper.

Published by John Wiley & Sons, Inc., Hoboken, New Jersey

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in

Library of Congress Cataloging-in-Publication Data:

1. GIS IN THE DIGITAL ECONOMY 1

2. INFORMATION SYSTEMS AND GIS 38

3. SUPPORTING BUSINESS DECISIONS 70

r (3) Branch-Related Campaigns / 80

5. CUSTOMER-FACING GIS: WEB, E-COMMERCE, AND

Mobile Services / 147

6. SPATIAL SYSTEMS DEVELOPMENT 169

7. THE VALUE OF INVESTING IN GIS 195

The Costs and Beneﬁts of GIS / 199

8. MANAGING SPATIAL DATA 227

9. ORGANIZATIONAL AND INDUSTRY ASPECTS 261

10. INDUSTRIES AND OUTSOURCING 295

11. ETHICAL, LEGAL, AND SECURITY ISSUES OF SPATIAL

12. GIS AND BUSINESS STRATEGY 355

The Future and Spatial Strategies in Enterprises / 381

Geographic Information Systems (GIS) support business decision-making by

support or take interest in a spatial project, customers were not consulted, or

Case Study Firms in GeoBusiness

environmental management, applied geography, and related disciplines. The

GIS IN THE DIGITAL ECONOMY

BUSINESS IN THE DIGITAL ECONOMY

2 GIS IN THE DIGITAL ECONOMY

Education at City College of San Franciso (CCSF, 2007). Universities have

The Rise in the Internet Platform for GIS

BUSINESS IN THE DIGITAL ECONOMY 3

TABLE 1.1 Number of Internet Hosts, 1984–2005

Year (January) Number of Internet Hosts

Source: Internet Systems Consortium, Inc. (http://www.isc.org/).

electronic commerce. Electronic commerce is the conduct of commercial

4 GIS IN THE DIGITAL ECONOMY

BUSINESS IN THE DIGITAL ECONOMY 5

6 GIS IN THE DIGITAL ECONOMY

the enterprise applications were often implemented by IT specialist groups

r A well-known estimate is that 80 percent of the development costs of a

BUSINESS IN THE DIGITAL ECONOMY 7

Attribute Tables Map

8 GIS IN THE DIGITAL ECONOMY

r Utilities. Attribute and boundary data on economic development, demo-

CASE STUDY: GLOBAL INTEGRATED OIL 9

r Environmental consulting. Data on environmental contamination, envi-

CASE STUDY: GLOBAL INTEGRATED OIL

10 GIS IN THE DIGITAL ECONOMY

Figure 1.3 GIO’s Energy Value Chain

Some notable examples of GIO’s spatial activities are the following:

CASE STUDY: GLOBAL INTEGRATED OIL 11

12 GIS IN THE DIGITAL ECONOMY

Figure 1.6 GIS in GIO’s Retail Market Planning in Africa

For transportation and storage, multi-layered maps show routes, depots,

CASE STUDY: GLOBAL INTEGRATED OIL 13