GPG231 Introducing Information Systems For Energy Management

GOOD PRACTICE GUIDE 231
Introducing information systems for

energy management
GOOD PRACTICE GUIDE 231
ARCHIVED DOCUMENT BEST PRACTICE

PROGRAMME
ETSU would like to acknowledge the assistance of Chenton Technology Management
and Hart Consultants in preparing this Guide.
ARCHIVED DOCUMENT
INTRODUCING INFORMATION SYSTEMS
FOR ENERGY MANAGEMENT
This Guide is No. 231 in the Good Practice Guide Series and is intended to provide
readers with a basic understanding of energy information systems. The technologies
and systems covered by this Guide can also be applied to services such as water and
compressed air.
Prepared for the Department of the Environment, Transport and the Regions by:
ETSU
Harwell
Didcot
Oxfordshire
OX11 0RA
and
The John Pooley Consultancy

Rectory Court
Great Witley
Worcestershire
WR6 6JP
ARCHIVED DOCUMENT
Crown copyright 1998
First published March 1998
LIST OF RELEVANT GOOD PRACTICE GUIDES
111. MONITORING AND TARGETING IN FOUNDRIES

112. MONITORING AND TARGETING IN LARGE COMPANIES
113. MONITORING AND TARGETING IN THE SEMI-MANUFACTURE OF NON-FERROUS METALS
119. ORGANISING ENERGY MANAGEMENT - A CORPORATE APPROACH
131. MONITORING AND TARGETING IN THE GLASS INDUSTRY
147. MONITORING AND TARGETING IN THE STEEL INDUSTRY
148. MONITORING AND TARGETING IN THE TEXTILES INDUSTRY
200. A STRATEGIC APPROACH TO ENERGY AND ENVIRONMENTAL MANAGEMENT
213. SUCCESSFUL PROJECT MANAGEMENT FOR ENERGY EFFICIENCY
215. REDUCING ENERGY COSTS IN INDUSTRY WITH ADVANCED COMPUTING AND CONTROL
217. CUTTING ENERGY LOSSES THROUGH EFFECTIVE MAINTENANCE (TOTALLY PRODUCTIVE
OPERATIONS)
Copies of these Guides may be obtained from:
Energy Efficiency Enquiries Bureau

ETSU
Harwell
Didcot
Oxfordshire
OX11 0RA
Tel: 01235 436747. Fax: 01235 433066. E-mail: [email protected]
Overseas customers please remit 3 per copy (minimum of 6) with order to cover cost of packaging and
posting. Please make cheques, drafts or money orders payable to ETSU.
ARCHIVED DOCUMENT
FOREWORD
This Guide is part of a series produced by the Government under the Energy Efficiency Best Practice
Programme. The aim of the programme is to advance and spread good practice in energy efficiency by
providing independent, authoritative advice and information on good energy efficiency practices. Best Practice
is a collaborative programme targeted towards energy users and decision makers in industry, the commercial
and public sectors, and building sectors including housing. It comprises four inter-related elements identified
by colour-coded strips for easy reference:
Energy Consumption Guides: (blue) energy consumption data to enable users to establish their relative
energy efficiency performance;
Good Practice Guides: (red) and Case Studies: (mustard) independent information on proven energy-
saving measures and techniques and what they are achieving;
New Practice projects: (light green) independent monitoring of new energy efficiency measures which
do not yet enjoy a wide market;
Future Practice R&D support: (purple) help to develop tomorrows energy efficiency good practice
measures.
If you would like any further information on this document, or on the Energy Efficiency Best Practice
Programme, please contact the Environment and Energy Helpline on 0800 585794. Alternatively, you may
contact your local service deliverer see contact details below.
ENGLAND
London North West South West
Govt Office for London Environment Team Environment and Energy Management Team
6th Floor Govt Office for the North West Govt Office for the South West
Riverwalk House Cunard Building The Pithay
157-161 Millbank Pier Head Bristol
London Water Street Avon
SW1P 4RR Liverpool BS1 2PB
Tel 020 7217 3435 L3 1QB Tel 0117 900 1700
Tel 0151 224 6401
East Midlands West Midlands
The Sustainable Development Team South East Regional Sustainability Team
Govt Office for the East Midlands Sustainable Development Team 77 Paradise Circus
The Belgrave Centre Govt Office for the South East Queensway
Stanley Place Bridge House Birmingham
Talbot Street 1 Walnut Tree Close B1 2DT
Nottingham Guildford Tel 0121 212 5300
NG1 5GG Surrey
Tel 0115 971 2476 GU1 4GA Yorkshire and the Humber
Tel 01483 882532 Sustainable Development Unit
North East Govt Office for Yorks and the Humber
Sustainability and Environment Team East PO Box 213
Govt Office for the North East Sustainable Development Awareness Team City House
Wellbar House Govt Office for the East of England New Station Street
Gallowgate Heron House Leeds
Newcastle-upon-Tyne 49-53 Goldington Road LS1 4US
NE1 4TD Bedford Tel 0113 283 6376
Tel 0191 202 3614 MK40 3LL
Tel 01234 796194
NORTHERN IRELAND SCOTLAND WALES

IRTU Scientific Services Energy Efficiency Office Business and Environment Branch
17 Antrim Road Enterprise and Lifelong Learning Dept National Assembly for Wales
Lisburn 2nd Floor Cathays Park
Co Antrim Meridian Court Cardiff
BT28 3AL 5 Cadogan Street CF10 3NQ
Tel 028 9262 3000 Glasgow Tel 029 2082 5172
G2 6AT
Tel 0141 242 5835
ARCHIVED DOCUMENT
CONTENTS
1 INTRODUCTION 1
2 DEVELOPING AN ENERGY 3
INFORMATION SYSTEM
3 DATA COLLECTION 6
4 DATA ANALYSIS TECHNIQUES 12
5 COMMUNICATION 20
6 ASSESSING ENERGY INFORMATION SYSTEM 23

NEEDS - A WORKED EXAMPLE
7 TOOLKIT 31
ARCHIVED DOCUMENT
INTRODUCTION
INTRODUCTION
The development of computers and information technology, together with
1
privatisation of the utilities, has changed the face of energy management
information. At the start of the 1990s, automatic data collection was considered
exceptional and only for those with significant capital to spend. However, the
development of metering technology necessary for the competitive energy market
means that many sites can now easily, and at low cost, access consumption data
every 30 minutes. We also have newly emergent bureau services using the latest
computer and communications technology.
This Guide looks at the basic elements of an energy information system and
discusses data collection, analysis techniques and communication methods. It
shows how to assess the requirements of a system with reference to a practical
example from an engineering site. Finally the Guide provides a series of tools that
will help you to assess the energy information system requirements of your own
company.
The systems and techniques covered can be applied to other services such as
water, industrial gases and compressed air. The Guide provides a basic introduction
to energy information systems but is not intended to provide definitive advice on
software, metering or other hardware.
Energy Information and Energy Management

For the purpose of this Guide an Energy Information System is defined as a
system for the collection, analysis and reporting of data relating to energy
performance that supports energy management.
An energy information system may be stand-alone, part of an integrated system or

a combination of several different systems.
For an engineering plant, an energy information system might comprise meters

connected to an existing data highway and accessed by dedicated data collection
software. The output could be used by a number of software systems, including
engineering, accounting and purchasing. For a commercial organisation, it might be
a software system, accessing billing data stored on an accounting system and
feeding into various management software modules. For a large multi-site
operation, it might be an external bureau service interfacing with an internal
management system.
An energy information system consists of more than meters and a PC. It also
includes all the organisational procedures and methods that allow it to operate.
Energy information systems can be hybrids that draw on external and internal
sources of data.
The prime function of the energy information system is the support of energy
management as part of the overall strategy of the organisation.
However, energy information systems are also needed to support environmental

management. For example, it may be difficult to become certified to ISO 14001,
Environmental Management Systems, without an effective energy information
system.
ARCHIVED DOCUMENT
1
INTRODUCING INFORMATION SYSTEMS FOR ENERGY MANAGEMENT
Technology
The impact of technology on energy information systems cannot be underestimated:
It is less than 15 years since the PC was introduced.
Most energy managers now have easy access to a reasonable

standard PC.
There have been significant cost and performance improvements in

metering and electronic data collection.
Many organisations have access to electronic mail and the Internet.
Whilst the cost of an entry-level PC has not decreased significantly, its capability for
data processing and presentation has increased considerably. We can collect data
at much lower costs, analyse larger data sets, and explore coloured graphic outputs
in ways we would not have thought possible five years ago.
Integration
Effective energy information systems are usually integrated with other systems:
hardware, software and organisational. Research carried out under the Energy
Efficiency Best Practice Programme shows that marginalisation is a major cause of
failure in energy management. Integration can help to avoid marginalisation and
can also provide more cost effective solutions.
Each case needs to be judged on its own merits. To develop an integrated solution,
you will want to talk to a range of people across the organisation, in particular, those
involved in the development of other information systems such as information
technology departments and management accountants.
How to Use This Guide

This Guide provides information and practical tools to help you establish or develop
the right energy information system for your company. It is an introductory
document that covers both buildings and industry. The principles that apply are
similar although some aspects differ in application.
Going Further
As you progress through the development of your energy management information
system, you will want to seek further advice and guidance on specific topics. This
may include setting up detailed systems, specific training or even using consultants
to advise on particular issues.
ARCHIVED DOCUMENT
2
DEVELOPING AN ENERGY INFORMATION SYSTEM
DEVELOPING AN ENERGY
INFORMATION SYSTEM
Most organisations will already have some level of energy information system, even
2
if it is not identified or managed as one. It should be appreciated that in a changing
working world, any information system will need to develop to meet its prime
objective - supporting management decision making.
The Operational Cycle

The day-to-day working of the energy information system can be illustrated by a
closed loop diagram.
Communication
(information to
understanding)
Data analysis Action

(data to information) (understanding
to results)
Data collection
The operational cycle
The operational cycle is based on four processes: data collection; data analysis;
communication; and action. These elements apply to any information system. The
function of the cycle is to turn data into action and thereby improve energy
performance.
Data, Information and Knowledge

Data are the raw materials of the information system. Examples include:
a listing of energy meter readings;
a record of maximum and minimum external temperatures;
a log of activity.
ARCHIVED DOCUMENT
3
In the raw form, data are of little use. Unfortunately, some energy managers
become totally immersed in data and see data collection and collation as their
primary task. To gain value from data they must be transformed into information
(used to support the knowledge development of all those managing energy) and
understanding (used to action energy savings).
Understanding
Information
Data
Energy information systems
System hierarchy
The system hierarchy shows how data, information and understanding are related,
how each is dependent on its predecessor and that information systems are needed
for support.
The three elements are also linked by a closed loop which leads from data through
information to understanding. It is also valid to work back from understanding to
determine the information needed and therefore the data collection requirement. In
practice, you will work in both directions to identify the optimum solution.
Whilst the prime purpose of an energy information system is to support energy

management, it has other valuable information outputs as well. For example,
improved allocation of energy expenditure may be of great value for internal
accounting purposes. This may influence budgets, product costings, etc. The ability
to provide such information provides you with a tradable commodity within your
organisation; tradable in the sense that in return for your information you can
access other information or resources that you need.
There is a wide range of solutions that can take you from data to information:
For an individual process it may be possible to use a simple graph or chart

to plot performance.
For a single site, a spreadsheet developed in-house might be appropriate,

working from manual meter readings and other relevant data.
For a site with a large number of meters, the solution may be a

sophisticated, real-time, data collection system linked to an analytical and
reporting system available over a network of computers.
For a complex, multi-site operation the solution may be a bureau-based

service, operated by an outside contractor, collecting data remotely and
providing reporting electronically.
Systems may be further sophisticated with an expert- or knowledge-based system

that provides advice and recommendations.
ARCHIVED DOCUMENT
4
DEVELOPING AN ENERGY INFORMATION SYSTEM
Low-tech High-tech
Sophistication
072170
072170
Examples
072170
072170
072170
072170
Visual reading Single PC based On-line metering,

and basic calculation system, specialised network, expert system
software
The range of solutions
At one time, information systems were limited by a lack of low-cost, effective

technology. We are now fortunate in having access to a wide range of metering and
information technology and Energy Managers must decide upon the most
appropriate technology to provide the solution required. It is critical is that the
information system adopted is appropriate for your organisation and is
owned by those that use it.
Two very different solutions are illustrated by these Good Practice Case Studies
(GPCS) from the Energy Efficiency Best Practice Programme:
GPCS 221, Monitoring and Targeting in a Hospital Laundry, shows how this
system at North Staffs Hospital Trust is based around a 1,000 software
package used on a single PC. Savings at the end of the first year were
10,300, giving a simple payback of around three months.
GPCS 330, A Comprehensive Energy Management Information System at

a Large Process Site, outlines how BP Chemicals introduced a
sophisticated, integrated system costing 5 million. This system achieved
2 million/year energy savings and avoided 2 million of capital
expenditure. Taking all savings into account it had a payback period of
1.5 years.
ARCHIVED DOCUMENT
5
3 DATA COLLECTION
Data collection is a fundamental part of any information system. However, it is all
too easy to become swamped with data. This happens when the data are seen as
valuable, rather than the analysis. When specifying the data collection system you
need to keep in mind the question:
What is the purpose of the system?
Data Types and Sources

Energy management data falls into three main categories:
consumption;
cost;
drivers.
Consumption
Consumption data are the most basic data required for energy management. They
are also essential for any environmental reporting. (For example, tonnes of carbon
dioxide generated by energy use can be estimated from energy consumption
multiplied by an appropriate factor. Similar calculations can estimate quantities of
sulphur dioxide or oxides of nitrogen). Consumption data can also cover other
utilities such as water and industrial gases.
For electricity and natural gas, the prime data source will be metering. Some fuels,
such as coal and oil, may be delivered in bulk and not metered on site. In these
cases, where metering is not provided, some alternative means of measurement is
needed. At one large manufacturing site, coal use is measured by the loader
bucketful!
Cost
Cost control is the main reason for most organisations to practice energy
management. Cost also provides a common language across various disciplines.
The principal source of cost information is the energy supplier, either from tariffs or
actual billings. If the information system is to support the organisations corporate
goals it must provide relevant financial information.
Drivers
A driver is any factor that influences energy consumption. Weather is the main
driver for most buildings and for most industrial processes, production is probably
the primary driver. Drivers are sometimes referred to as variables or influencing
factors.
ARCHIVED DOCUMENT
6
DATA COLLECTION
There are two main types:
Activity drivers. An activity driver is a feature of the organisations activity

that influences energy consumption. Examples include: hours worked,
tonnes produced, number of guests and opening hours.
Condition drivers. A condition driver is one where the influence is not

determined by the organisations activity but by prevailing conditions.
Examples include: weather, condition of the raw material and hours of
darkness.
Data probably already exists for production and other business drivers, but the A metal finishing contractor in the
quality of this data should not be taken for granted. It needs to be checked. automotive sector delivered
processed parts in batches of
Typically, collection of activity data will be from internal sources. Weather data,
1,000 as part of a Just in Time
usually degree days, may be externally provided or supplied internally using
supply chain. Observed variations
dedicated equipment or a Building Management System (see Fuel Efficiency
in energy consumption ratios led to
Booklet No 7, available from the Energy Efficiency Best Practice Programme, for
a check on the actual batch sizes
further details of degree days).
which were found to range
Single and multiple sites between 850 and 1,200. This
illustrates the need to verify the
This is the area where we see some of the greatest differences between industrial accuracy of the data input to the
and buildings-based systems. For example, a retail shop chain may have hundreds energy information system.
of sites, each with quarterly billing. Alternatively, a large industrial site may have 20
incoming meters with half-hourly readings and a large number of internal sub-
meters.
One important point about multiple sites is that each site will normally have its own
utility supply and associated billing. This means that utility invoices should already
be available for each site and provides a good starting point for the energy
information system.
Meters and Metering Periods

Main meters
A main meter is the supply meter provided by the utility or meter operator for the A large car manufacturing site has
purposes of charging for supply. (It may also be called a fiscal meter or a statutory 20 main electricity supply meters
meter.) Typically, there will be one main meter for each utility supply. However, in which are all read remotely by the
the case of larger sites, or complex supplies, there may be multiple main meters. utility at half-hourly intervals. The
Main meters should normally meet an appropriate British Standard and the Energy Manager accesses this
utility/regulators code. half-hourly data to prepare weekly
energy reports.
Main
meter Main meter
Sub-metering
First level
sub-meters
Second level
sub-meters
Meters and sub-meters
Sub-meters
A sub-meter is a meter installed after the main meter. Extensive sub-metering is
used in organisations where departmental/process cost accountability is required.
(In these circumstances, the prime purpose of the meters is cost accountability with
energy management information as a by-product.)
ARCHIVED DOCUMENT
7
A feature of sub-metering which must be appreciated is that for a totally sub-

metered supply the sum of the consumption of the sub-meters does not normally
equal the consumption of the main meter. This is due to the accuracy bands of the
meters involved. For example, the main meter might be 0.5% with the sub-meters
at 2%. In the case of gas, the main meter may be pressure- and temperature-
corrected and the sub-meters not.
More meters and more frequent reading?

Code 5 is a Code of Practice from With modern metering systems, any site with an electricity Maximum Demand (MD)
the Office of Electricity Regulation over 100 kW can have Code 5 metering which will provide half-hourly electricity
(OFFER), detailing the requirements consumption data. Without Code 5 metering many sites only have monthly meter
for half-hourly metering systems. As readings from their utility company.
part of the progress to full electricity
supply competition in 1998, it is Would fewer meters read more often provide better data/information/
proposed that all sites using 100 kW understanding? An industrial site might get more useful data from 5 meters read
or above will be fitted with Code 5 every half hour than from 50 meters read once a month.
metering.
For a local authority, where the cost of sub-metering may not be justified, half-hourly
metering from main meters may be the only cost effective way to investigate
individual site performance.
Metering periods
The metering period is the time between meter readings. The reporting period is
the time between energy information reports.
For a Code 5 site, the metering period would be 30 minutes, but the reporting period
could be weekly or monthly. The metering period is often determined by the
technology used. Many systems not using on-line metering use a monthly period.
A month may be 28, 29, 30 or 31 days, or even 35 days when a five-week

accounting period is called a month! The week is a better period to use for a number
of reasons:
it has a universal definition - 7 days;
if a reading is missed, it is only one of 52, rather than one of 12;
suitable sets for data analysis can be generated quicker;
the week also allows the system to be more responsive and timely.
The major disadvantage of weekly reporting is the increased amount of data

collection and entry. In practice, this needs to be balanced with the improved quality
of information delivered.
The use of on-line metering provides greater flexibility in the choice of metering and
reporting periods.
Data Collection Methods

The complexity of data collection depends on the:
number of data collection points;
method of data collection;
frequency of data collection.
Physical distance is also relevant. Single sites such as an international airport, a

large hospital, or a chemical works, may have distances of several kilometres
between remote meters and the central data system.
Consumption data comes from two main sources: invoices and direct metering.
Collecting data from invoices

Double processing of energy data Invoices are the major source of energy data for many buildings-based systems.
should be avoided. A common Typically, for a large property portfolio, there will be separate gas, electricity and
example is where accounting and water accounts for each site. Processing this invoice data can present a significant
energy management functions are work load in a large organisation. However, much of the processing effort is
working in isolation. required for accounting purposes. In such cases, integration between accounting
and energy management is critical.
ARCHIVED DOCUMENT
8
DATA COLLECTION
If sites have the appropriate metering then the utility company will use actual meter
readings to compile the invoice. However, problems exist with smaller, remote, or
difficult to access sites where the utilities often use estimated readings. Such
invoices have little value in energy management terms. In these circumstances, you
need an alternative source of data.
As part of their overall function, many energy information systems will be required to
validate invoice data using metering data. There are varying definitions of invoice
validation. These range from simple checking of the arithmetic on the invoice to
software packages which use an estimation or prediction of the consumption and
compare this with the invoice. Full invoice validation must include the following:
confirmation of site and tariff details;
checking of arithmetic;
comparison of utility reading with organisations own reading.
Collecting data from meters

There are two basic methods of reading energy meters: manually or electronically.
Manual (or visual) readings are taken by staff recording data from either dials or
digits on the meter. These give a quantitative measure, in units appropriate to the
utility, such as kWh for electricity and m3 for water. In some cases, a metering
multiplying factor may be needed to convert the reading into the appropriate units of
consumption.
Data can be collected by noting down the figures on a collection sheet or by using
a hand-held electronic data collection unit (DCU). Data collection units often include
a bar code reader to identify the meter and are pre-programmed to check the validity
of the data.
The most common electronic meter is the pulse output meter. This meter generates
electronic pulses proportional to the quantity measured. A pulse counter is required
to make use of the readings. It is also necessary to know the ratio of pulses to
consumption. Most meters which have an electronic output also have a visual
output which can be used to validate the electronic data.
Meters with electronic output comprise either a basic meter which simply generates
output pulses or an intelligent meter which can store data for reading at a time
convenient for the meter reader. An advantage of the intelligent meter, over the
basic pulsed output meter, is that stored output can be used to cover data
transmission problems.
Having measured the consumption and obtained a reading, the information needs
to be transmitted to the next stage in the information system. Visual readings may
be transmitted on paper or electronically from the data collection unit. They can
also be entered into a PC and then transferred electronically. Whereas manual
readings must be manipulated by hand or entered into a PC, electronic signals can
be transmitted directly to the data collection unit.
Electronic options include:
direct hardwiring;
modem;
mains-borne carrier techniques;
links to data highways;
low power radio systems.
Each method has advantages and disadvantages: the use of a modem may reduce
wiring costs, but incur telephone charges; direct hardwiring may give secure,
reliable, low-cost transmission but it also adds considerably to installation costs.
ARCHIVED DOCUMENT
9
Low-tech High-tech
Sophistication
072170
072170
072170
072170
072170
072170
Visual reading Direct electronic

with manual reading with
records data storage
Sophistication of consumption data collection
Energy cost information

Cost information normally comes from the utility supplier. The time frame of the cost
information should be considered with the consumption data collected. This is
usually more of a problem with electricity when the actual unit rates charged may
not be known until the end of the billing period. At a simple level, if energy is
supplied at a fixed unit rate, cost information can be generated as the consumption
data is collected. One solution to the problem of unknown costs is to use an
estimated or past average unit cost.
If you use estimated costs, how do you handle differences between estimated and
actual costs? One solution is to use consumption figures for immediate energy
management use and then provide financial information separately later.
Data on drivers
Data on activity or production-related drivers will usually already be available and
often in electronic form as a part of normal business management information
systems.
External climate is an important driver and is typically recorded as degree days.

Degree days are derived from the maximum and minimum external temperatures
over a 24-hour period. Many Building Management Systems can collect data for
degree days, it is also possible to buy dedicated degree day monitors. Some sites,
for example, airports and universities, may be already collecting daily maximum and
minimum temperatures for other purposes. Degree days are more applicable to
traditional buildings with heating and natural ventilation, and less applicable to
integrated design buildings with full air conditioning and high internal heat gains.
The collection of driver data will be dependent on other/existing information

systems, in contrast to energy consumption data which will probably be only
collected by the energy information system.
ARCHIVED DOCUMENT
10
DATA COLLECTION
The cost of data collection

Whichever data collection option you follow, be it low or high technology, it comes at A local authority site had a modem
a price. There is no point in spending more money to collect data than can be link for monitoring and control. The
saved by the useful application of that data. telephone and maintenance costs
of the system were found to be
Points worth considering about the costs associated with data collection include: 50% of the energy expenditure.
for a given energy type, the cost of data collection is not normally related to A large industrial manufacturing
the level of consumption. It costs the about the same to collect data from a site was found to be taking over
meter recording 50 of energy as it does for one recording 5,000 of three weeks each month to collect
energy; the data from visually read meters.
time and money are both data overheads; Clearly neither are optimum
investment in data collection should lead to reduced running costs; solutions.
manual meter reading is an identifiable running cost;
all energy information data collection systems have ongoing running costs.
Data Quality
Data are the raw materials of the energy information system and good quality output
is dependent on good quality raw materials.
Good quality data should not be confused with high accuracy data.
For energy management purposes, the level of data accuracy required is not
excessive but the data must be repeatable and dependable. Time spent ensuring
the quality of the base data in any information system will not be wasted. Some
analytical techniques will identify suspect data points but it is better if they are not
included in the first place.
The data in any energy information system should be able to withstand the rigours
of a formal audit process. All too often it is assumed that energy information
systems are self auditing - this is not the case.
Some examples of the need for good data quality:
The electricity consumption in a hospital boiler house was read as one-tenth

of actual consumption for over five years. This was because the first reading
was not checked properly and all subsequent readings were validated against
the original.
A retail organisation using invoice data to track energy use discovered that
some of its sites had only received estimated readings over the past three
years.
A local authority had confused the energy data for two sites. This was only
revealed when one site was closed!
ARCHIVED DOCUMENT
11
4 DATA ANALYSIS TECHNIQUES

Data analysis is a vital process in the transformation of data into information. The
desired information output will influence which analytical techniques are used.
However, during the development processes, data availability may influence the
analysis techniques considered. Data analysis normally develops through two
stages.
Stage 1. Looking only at energy. For example, converting meter readings

into consumption figures, comparing the current period of consumption with
the same period the previous year. Stage 1 techniques are often used at
the outset based on historical invoice or consumption data.
Stage 2. The use of energy data and drivers. The results of this should
prompt questions: Why does that happen? Should that happen? Is that
what we expect? Can we do better? How well do we compare?
There are two ways in which analysis techniques are applied:
Routinely. To provide the regular output of the system. Routine analysis

can often be automated, or carried out by someone without specific energy
management skills.
Investigative. To investigate the process rather than report upon it. This
type of analysis can be applied periodically to review the effectiveness of
the system or it may be used to follow up exceptional results arising from
routine analysis. This application requires a level of energy management
skill.
As this is an introductory Guide, the details of applying analysis techniques are not
covered. However, the scope and application of the majority of techniques in use
are discussed. Some techniques are more suitable than others, depending on the
organisation or activity. Certain techniques have been developed for specific
applications and all have limitations on their use. You need to identify these issues
when considering which techniques to use.
Data Volume
The volume of data available will have an influence on the techniques used. For
example, the availability of half-hourly electricity data has lead to the development
of profile analysis and contour mapping techniques. Such techniques require a high
volume of data.
Statistical techniques need a minimum size of data set, typically between ten and 20
points. (The set comprises a value for the energy and the driver for each point.) If
an analysis of the relationship between heating and weather is required, you can
encounter problems using monthly data. This is because the heating season is
usually less than ten months, so the data set collected is too small to apply the
technique at the appropriate level of confidence. One solution is to use the data
from two heating seasons, this requires the system to be identical in operation for
those two seasons. An alternative is to use weekly data. This enables more than
one suitable data set to be obtained from a single heating season. (However, this
will also increase the amount of energy data to be collected and require the use of
weekly degree days.)
ARCHIVED DOCUMENT
12
DATA ANALYSIS TECHNIQUES
In general, the more data collected, the wider the range of analysis possibilities,
although an increase in data collection usually leads to increased overheads.
Another potential problem is paralysis by analysis. Unlike drowning in data, this
is when the analysis becomes the primary function and analysis leads to further
analysis, all at the expense of effective energy management.
Selecting Analytical Techniques

Not every organisation will need to use every technique, and some techniques may When selecting techniques, key
not be economically viable. questions are:
As the system develops, the techniques used may change and become more Which techniques are
sophisticated. However, it is important to remember that there may be considerable appropriate to my situation?
inference that can be made from existing data.
What do others use in my
The first step is always to study available data before developing the system. sector?
Many practitioners have been able to demonstrate significant potential for Which techniques can I use to
performance improvement simply by looking at existing data in a different way. As benchmark my performance?
part of the regular review of a system, you should try different analytical techniques
to see what they show. What do I want/need from the
analysis?
For any set of data there are many techniques that can be applied. These range
from simple comparison techniques, through line fitting, to sophisticated statistical How much data is required?
methods. The techniques selected will depend on your aims. First understand the
techniques and then select those suited to your use. Do I have that data? If not, can
it be obtained economically?
This Guide outlines ten data analysis techniques that can help to present and
interpret energy and driver information.
Analysis technique Buildings Industrial Description

sites
Normalised Performance benchmarking against buildings of similar type.

Indicators (NPI)
Specific Energy simple industrial process benchmarking.

Ratio (SER)
Current and Past comparisons against previous energy performance.

Comparison
Trend Line graphical display of energy use against time.
Profiles to show consumption patterns over specific time

periods.
Contour Mapping 3-D way of displaying of energy

profiles.
Lines of Best Fit for approximating simple mathematical relationships

between energy consumption and key drivers.
Variances to show deviation from anticipated energy

performance.
CUSUM CUmulative SUM of variances from standard

performance - useful to identify changes in the
pattern of energy use.
Control Charts using predetermined control limits to alert exceptions

to planned performance.
ARCHIVED DOCUMENT
13
Normalised Performance Indicator (NPI)

90
Three hospitals compared
against performance Poor
80
yardstick figures Fair
70
Good
60
GJ/100 m3/year
50
40
30
20
10
0
Hospital A Good Hospital B Poor Hospital C
Normalised Performance Indicator (NPI)
The Normalised Performance Indicator (NPI) is a technique for buildings, usually

calculated annually. It provides a yardstick figure of energy consumption as kWh
per m2/year (in the case of the health sector, it is more usually volume based i.e.
per m3). The calculation requires a total annual energy consumption and a floor
area or volume. This gives the Performance Indicator. It can then be normalised
by reference to tables covering operating hours, weather, etc. NPIs can be
produced for total energy; energy type (gas, electricity, oil etc.); or by use (heat,
lighting, air conditioning etc.).
The NPI allows the comparison of buildings of a similar type. This comparison can
be used within the organisation, against other organisations or with benchmark data
from publications produced by the Energy Efficiency Best Practice Programme.
As this technique is based on floor area, it is important to know what floor area
figures you are using. Are they accurate? Are they for the total floor area or
occupied floor area? This is particularly important when making comparisons.
The NPI technique can be refined by using a moving annual calculation, on a

monthly basis. This extension to the technique gives monthly output, allowing
regular review against annual targets. It can also be used to provide a trend line.
Specific Energy Ratio
72 560
Raised production level
70 leads to a fall in SER Production level 550
540
68
530
66
Tonnes/week
Therms/tonne
520
64
510
62
500
60
490
58
480
56 470
54 460
15 16 17 18 19 20
Week number
Specific energy requirement for glass melting
Specific Energy Ratio (SER), also known as Specific Energy Consumption (SEC),
is a technique used in industry. It is simply the energy used divided by an
appropriate production measure (i.e. a driver). It can be calculated for any fixed time
ARCHIVED DOCUMENT
14
period, or by batch. SERs need to be treated with care because their variability may
be due to factors, such as economies of scale or production problems, rather than
energy management.
There are many process benchmarking schemes based on SER and their ease of
use makes them attractive to many companies. SERs feature widely in the Energy
Consumption Guides produced under the Energy Efficiency Best Practice
Programme.
Some practitioners are strongly averse to SER, regarding it as too simplistic and
flawed. However, it has the value of being easy to calculate, understandable and
straightforward to communicate, making it widely used in many industries.
Current and Past Comparison

12,000
Past year
Current year
10,000
Annual electricity cost ()
8,000
6,000
4,000
2,000
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Current and past energy expenditure
This approach is suitable for buildings and industry. It is usually in a graphical format
where a bar or column chart is used to compare the data from the current period with
a similar previous period. A tabular form of this comparison can also be used with
a quantity or percentage figure for the difference. It is useful for monitoring year-on-
year changes and cyclical patterns, and can also be used for daily and weekly
profiles. The technique can be applied to energy-only data, or energy/driver data.
Trend Line
This approach is suitable for buildings 107,500
and industry. Most energy managers
are interested in the underlying trend of
consumption or cost and trend lines 107,000
are a graphical way of showing this.
Annual electricity costs ()
Typically, the trend line will be the trend

106,500
of the data series over time. At its
simplest, it is a line graph of the data
for each period. 106,000
A more refined application of the

technique is to use moving annual 105,500
totals or averages. This approach is
useful since it reduces seasonal
105,000
influence and allows other influences
to show through. Since a trend line
can be produced from time-related 104,500
Dec Jan Feb Mar Apr May Jun
energy data alone, it is a common
12 months ending
technique to use at the early stages of
investigating energy consumption.
Moving annual costs
ARCHIVED DOCUMENT
15
Profiles
9,000
8,000
7,000
6,000
5,000
kWh
4,000
3,000
2,000
1,000
0
1 Feb 8 Feb 15 Feb 22 Feb
Date
Site electricity profile
This approach is suitable for buildings and industry. Profile analysis has gained in
popularity with the wider availability of half-hourly electricity metering. Profiles for
the day, week, month or year can be prepared. The technique can be expanded by
the comparison of current and past profiles, average profiles, or checking against
limits set on a base profile. Profiling is particularly valuable where the pattern of
consumption is repeatable. As well as being used visually, profiles can be used
arithmetically. For example, subtracting the data from two profiles and displaying
the result either as a table or a chart.
Contour Mapping
28 Feb
26 Feb
24 Feb Electricity
consumption (kWh)
22 Feb 200 - 300
20 Feb 100 - 200
0 - 100
18 Feb
Date
16 Feb
14 Feb
12 Feb
10 Feb
8 Feb
6 Feb
4 Feb
2 Feb
00:30
01:30
02:30
03:30
04:30
05:30
06:30
07:30
08:30
09:30
10:30
11:30
12:30
13:30
14:30
15:30
16:30
17:30
18:30
19:30
20:30
21:30
22:30
23:30
Time of day
Electricity contour map
This is approach is suitable for buildings and industry. The contour map offers a
more pictorial use of profile information. Here, half-hourly data, typically for a month,
is displayed as a multi-coloured contour chart. This provides a very easy way of
viewing 1,400 data points (30 days x 48 half-hours).
ARCHIVED DOCUMENT
16
Lines of Best Fit
3,000
2,800
2,600
2,400
2,200
kWh
2,000 Line of best fit equation

y = 4.91x + 242.58
1,800 Correlation coefficient
R2 = 0.82
1,600
To improve clarity
1,400
chart has not been
continued to zero
1,200
1,000
300 350 400 450 500 550
Units produced
Line of best fit - energy vs. production
This approach is suitable for buildings and industry. Understanding how drivers
influence energy consumption is one of the most important energy management
techniques. The starting point is to plot energy and the selected driver on an X-Y
chart (a scattergram). Usually the Y axis is chosen for energy and the X axis for the
driver.
From process knowledge, you need to assess if a straight line is a reasonable

relationship to expect between the driver and energy use. (For example, we can be
reasonably confident of this with space heating or metal melting.) If this is a
reasonable assumption, we can produce an equation by inserting a straight line in
the scatter, i.e. a mathematical model, describing the relationship between driver
and energy. It may be necessary to experiment with several different drivers before
a suitable relationship is determined.
For a given value of driver on the X axis, reading the value on the Y axis, by way of
the straight line, will give an expected value for energy. (The value of X can be used
in the derived equation to give a value for Y). Note that care should be taken if the
values of driver are outside the range of the chart as confidence in the model is
questionable outside this range.
The technique can be employed in its simplest form with pencil, graph paper and
ruler. (This is a good way of understanding what is involved.) A statistical
calculation, known as regression by the method of least squares, can be used. An
effective way to explore this technique is to use a PC spreadsheet that includes the
function.
Most spreadsheets will include a function that will identify a line of best fit and
calculate its equation:
Excel - create an X-Y chart and use the commands: Chart/Insert/Trendline.
Lotus - from a table of data use the commands: Range/Analyse/

Regression.
This technique is also included in most commercially-produced Monitoring and

Targeting (M&T) software. When using the technique it is important to use good
quality data. Any suspect points should be checked out and, if necessary,
eliminated from the analysis.
When first applying this technique, some users are concerned about data scatter
and correlation coefficients. (Correlation coefficient is a measure of how well the line
represents the data. When using a spreadsheet it can be calculated automatically).
Typically, a well-controlled process will exhibit a correlation coefficient of 0.9 or
higher. Do not be disappointed if your process does not achieve this.
If the underlying relationship between the energy and the driver is correct,
then a large scatter (low-correlation coefficient) indicates poor control and
large scope for improvement.
ARCHIVED DOCUMENT
17
The line of best fit established by the technique is called the Standard Line.
Additional analysis can be undertaken to determine a Target Line - this represents
a planned improvement on standard performance.
Standard statistical texts plus Fuel Efficiency Booklet No. 7, Degree Days, and Good
Practice Guides on Monitoring and Targeting all have examples of line of best fit.
Further information on line of best fit can be found in a range of text books.
Variances
400
300
Energy more than

200 prediction/standard
100
Variance (kWh)
-100
Energy less than

-200 prediction/standard
-300
1 2 3 4 5 6 7 8 9 10 11 12
Week number
Weekly variance chart
This approach is suitable for both buildings and industry. Having established a
standard or expected level of energy consumption, a simple and useful routine
analysis is to calculate the variance, or deviation, of actual consumption from the
expected or standard consumption. This variance can be charted and/or tabulated.
Variance analysis is often popular with accounting-based users as they show
parallels by comparing actual expenditure with budgeted expenditure.
CUSUM
This approach is suitable for buildings
200 and industry. CUSUM is the
CUmulative SUM of the differences
150 from standard or predicted
performance. The technique
Change originates from quality control and,
Cumulative variance (kWh)
100
following development work, has been
widely applied to energy management.
50
Change
CUSUM can be simply calculated,
using a spreadsheet, but needs to be
0
displayed as a graph or chart to gain
Change full benefit of its use. A characteristic
-50
Change feature of a CUSUM chart is that it
usually produces a series of straight
-100 lines, representing steady
performance, separated by sharp kinks
-150 indicating events which change the
13 14 15 16 17 18 19 20 21 22 23 24 underlying pattern.
Week
There are two important applications of
CUSUM chart CUSUM in energy management
information systems:
ARCHIVED DOCUMENT
18
Identifying and quantifying the impact of changes in the pattern of energy

use.
Identifying the time of the last change in pattern, enabling data to be

identified which can be used as a basis of control.
The quality of CUSUM analysis depends on the quality of the data used and the
relevance of the baseline calculation. Changes in baseline equation can produce
significant changes to the shape of the CUSUM line. Once correctly established,
CUSUM provides a powerful technique to investigate and monitor performance.
Control Charts
20
15
Upper control limit (+10%)

10
5
Variance (%)
-5
-10
Lower control limit (-10%)
-15
-20
1 2 3 4 5 6 7 8 9 10 11 12
Batch number
Control chart
This approach is suitable for buildings and industry. Control charts also come from
quality control and anyone in manufacturing familiar with Statistical Process Control
(SPC) will have used a control chart. For a control chart, a baseline value is
required. This value can be derived by using CUSUM to identify the relevant base
data. Control limits are then established around this baseline; these limits may be
arbitrary or can be developed from an analysis of the process. The limits can also
be used to provide an alert or alarm status, for example 5% and 10%. This
approach can be particularly useful if a large number of meters are being monitored
and exception reporting is required.
ARCHIVED DOCUMENT
19
5 COMMUNICATION
Having collected and analysed the data, it must be communicated to complete the
transformation from data to understanding. If an energy information system is to
fulfil its prime objective - to support energy management - effective communication
is a critical part of that system. There is little point in collecting, collating and
analysing data for storage purposes only. Equally if communication is little more
than a reiteration of the base data, the system has not succeeded in adding value
to the data.
Communication
Action
Information Added value Direction Action
Action
Communicating data
Typically, communication falls into one of three categories:
Regular - typically, the weekly or monthly report. Issued on a time basis

rather a need basis.
Exception - produced when something has gone wrong - or right!
Ad-hoc - initiated by request or as the result of investigation. Possibly

produced to coincide with a relevant event or activity.
Identifying Information Needs

It is essential to discover who needs what information to support their role in the
energy management process. This means finding out what people need rather than
making assumptions. It should be clear that the information needs vary throughout
the organisation. Consider an engineering factory: What information does the
managing director need? Is this the same as the works engineer or the production
scheduler? Consider a hospital: What information does a unit manager need? Is
this the same as a ward nurse, or unit engineer?
ARCHIVED DOCUMENT
20
There are two basic questions that you need to answer. There are generic answers
but the specific answers will depend on individual circumstances.
Question: Who needs energy information?

Answer: All those individuals that can influence energy performance.
Correctly identifying the recipients of energy information is essential for
communication to be efficient and effective. Therefore who needs to know? The
answer, though very general, is anyone and everyone who can have an impact. For
any organisation this will cover a wide range of people.
It could be argued that, for most organisations, everyone needs to receive some
energy information. With the growth of environmental management, we also have
to consider communication with stakeholders. Stakeholders include shareholders,
employees, regulatory bodies, interested groups and members of the public. To
satisfy this need, the external reporting/communication requirements for most
organisations will increase.
What should be immediately apparent is that information needs vary widely. At one
extreme it may be a single key statistic, at the other a comprehensive and detailed
report.
Question: What information do recipients need?

Answer: The minimum necessary to enable them to improve energy
performance.
Not everyone needs to know everything. The information that is communicated
should be the minimum required for achieving results. It should also be information
that recipients can easily assimilate. This may mean training, it may mean that the
style and nature of output needs to be tailored.
One approach to consider is providing sample outputs for comment. Many of the
commercial bureau services start by offering a wide range of reporting outputs to
allow the user to make a choice. They then optimise the system to a limited rage of
outputs selected by the customer.
Annual Monthly Weekly Key Exception

A major financial institution was
report report report indicators report
reporting energy consumption to
its staff in consumption terms
Chief executive only, i.e. kWh. As part of a staff
survey, it became apparent that
they felt a need for the
Accountant information to be presented in
financial terms as well. The
Department heads opinion was that, as finance was
the companys core activity,
anything not explained in
Purchasing financial terms had little
relevance.
Supervisors
Workforce
What information do recipients need?
Timing of Communication
There are two key time-related aspects to the issue of communication:
Speed - The value of information is time dependent. Information delivered

at the right time has a high value. Information delivered too late, when little
can be done, wastes time in being assimilated and can discredit the system.
Frequency - Communicating too often introduces information overload. It

is commonly accepted in todays work environment that managers have
access to too much information, rather than too little. Conversely sending
reports and information too infrequently can mean that interest is lost.
ARCHIVED DOCUMENT
21
These two examples show how Communication Methods

established communication
methods are often the most A number of communication methods are available. The most commonly used are
successful: tabulated figures and graphs/charts. In many cases, the chosen method is pre-
determined by the software used or organisational practices. It is important that the
At a university it was noted that options chosen by these limitations match the needs of the organisation.
water consumption had
increased significantly. The Low-tech High-tech
Facilities Manager decided to
Sophistication
send an e-mail with the figures
and costs around the campus.
It took him less than ten
minutes. Within a month water
consumption had fallen by
30%.
At a large car plant, use was

made of the electronic notice
board system. (In practice
similar to an on-site teletext Face to face Customised Presentations Electronic,
reports on demand
system.) This used a simple
graphic to indicate energy cost
performance. Communication methods
For some people, you may need to deliver information in an exact form to their desk;
for others you may need only to give them access to the system. The best methods
to choose are the ones which achieve results. This will vary from site-to-site and
from year-to-year. What works for one group of people may not work for others.
ARCHIVED DOCUMENT
22
ASSESSING ENERGY INFORMATION SYSTEM NEEDS - A WORKED EXAMPLE
ASSESSING ENERGY
INFORMATION SYSTEM NEEDS -
A WORKED EXAMPLE
In this Section the example of an engineering site is used to show the process of
6
assessing your companys information system requirements. The process has five
steps:
Defining
the system
Gathering
information
Selecting
analysis techniques
Assessing
existing systems
Taking
action
The example engineering site has an annual energy bill of a little over 1
million/year. The site consumes electricity, gas and water plus diesel for its transport
fleet. The company uses a software package for Monitoring and Targeting (M&T)
and its information system is based upon monthly meter reading and reporting.
Defining the System

As the function of the system is to support energy management, it should be
apparent that the information system must exist as a part of a structured
approach to energy management. If such a structured approach does not exist
within the company, this needs to be addressed first.
If the company has a structured energy management approach you can proceed to
defining the requirements of the energy information system. You should set clear
quantifiable objectives, taking into account the present systems and levels of energy
efficiency and the resources you have access to. This system definition is not a
technology wish list: it needs to be a clear set of requirements and objectives for a
system to support energy management, both technology and people.
To help frame the system definition it is important to look at the energy savings
potential in your organisation.
ARCHIVED DOCUMENT
23
The following table shows utility spend and savings estimates for our example of an
engineering site.
Savings potential
Utility Annual % of % savings Savings
spend k total spend potential k
Electricity 510 50.0 4 20.4
Gas 220 21.6 6 13.2
Water 180 17.6 15 27.0
Diesel - transport 110 10.8 10 11.0
TOTAL 1,020 100 71.6
Total savings/annual spend 7%
The spend figures were taken from the M&T system. The savings potential figures
are consultants estimates. In this example, it can be seen that water presents the
greatest savings opportunity; does the existing system support this area? Also,
electricity accounts for half the total expenditure; again does the system support
this?
Another part of defining the system is to ask people what they want from the system.
This will be of limited value if they do not understand energy management, although
you can find out what form and style the information output should take. Answers
might range from a full written report to occasional on-line access. Form and style
will also be influenced by the custom and practice of the organisation.
Gathering Information
You will need to gather information about the companys existing metering and
invoicing data and the methods of communication that are available to you.
This data gathering should include:
raw data - meter reading and invoices;
analysis - calculations and assessments;
communication - reports and other outputs.
You should also consider drawing up a flow chart for all these items. Questions you
should be asking include:
What raw data is already available?
Which analysis processes are used and why?
Where is data and information stored?
How is information communicated?
Do people get what they want?
Is information delivered on time?
Is the information both of good quality and reliable?
ARCHIVED DOCUMENT
24
Metered supply - main meters
Utility Number ID/ Type Reading frequency

of meters location A = on-line metering Q = quarterly; M = monthly
B = (Basic) manually read; W = weekly; D = daily
Other - specify 30 = 30 minutes
Other - specify
By By
utility organisation
Electricity 1 M186 A 30 W
Main gate
Gas 2 M0875 B M W
M67899 B M W
Gas house
Water 1 12567 B M W
Entrance road
Total 4
You may find it advantageous to include meter numbers (M numbers) and

postcodes, as this data is required when seeking competitive supply quotations.
The example shows the possibility of half-hourly electricity readings which are not
being used.
Not all utilities will be metered at point of supply. For our example site, the only bulk
supply is road diesel delivered to the underground storage tanks. In this case, the
consumption of the diesel is metered by way of the fuel pump when vehicles re-fuel.
Accordingly it is possible to get consumption data by vehicle.
Bulk supply - unmetered

Utility Method Frequency Consumption
of charging of charging metering
Road diesel By delivery By delivery At point of use
The next area to consider is the processing of invoices. For our example company
this will be straightforward. This is the area where large buildings-based
organisations will differ significantly from our industrial example.
Invoice tracking
Detail Gas Electricity Water Total
Invoices per month 2 1 1 4

(48 per year)
Paper or electronic? P/E P P P
Estimated readings? Y/N Y N Y
Direct debit payment? Y/N Y Y Y
We can see that with four paper invoices a month (48 a year) invoice processing
should not be a major task. The next part of invoice tracking is to produce a flow
chart tracking the progress of the invoices, and copies, around the organisation.
This physical tracking is a very important part of the review of multi-site buildings-
based systems.
ARCHIVED DOCUMENT
25
Day 1
Invoice arrives
Incoming accounts
Day 1
Logged in
Day 2
Copy to engineer
Day 6
Scheduled for payment
Filed in accounts
Day 14
Paid direct debit
Invoice flow chart
Here we see that the engineer is included on an information only basis with
payment automatically taking place. If the engineer is to be part of the invoice
validation process, then he/she needs to respond quickly to meet the direct debit
payment schedule, and be part of the main flow line.
Moving on from the supply, it is possible that sub-meters have been installed. These
should be logged in a similar way to the main meters.
Sub-meters
Metered Number Type Reading frequency
utility of sub-meters A = on-line metering Q = quarterly; M = monthly
A = on-line, X = not read
Electricity 5 B W
Gas 0
Water 0
Diesel 2 B Tank filling
Total 7
For our example site we have five electricity sub-meters but no water sub-meters,
yet the greatest potential for saving is with water. Are more water meters required?
If we are to correlate energy use against production or other drivers, we must

identify and gather information on the factors that have an influence on energy
consumption.
Energy drivers
Driver Process affected Source of data
and energy type
Tonnage Furnaces and machinery - Production control system

electricity and gas
Weather Heating - gas and electricity Energy management

degree days magazine
Mileage Fuel consumption - diesel Transport log books
ARCHIVED DOCUMENT
26
In our example, the site has chosen typical drivers. We cannot ascertain how
relevant they are without additional site knowledge.
The next area of interest is communications. Try to identify all forms of

communication that might be relevant.
Communication
Report Frequency From To
Company Annual Energy Manager Managing Director

Energy Report and
Department Heads
Energy Indicators Monthly Energy Manager Department Heads
Budget Report Monthly Accountant Relevant

Department Head
Fuel Economy Monthly Fleet Manager Accountant
In our example, two of the communications did not originate with the manager
responsible for the energy information system. These are a financial budget report,
produced by the accountant and a fuel economy report produced by the fleet
manager. Although not produced by the energy information system, these reports
are communicating energy information. Should they be integrated? Do they agree
with the output from the energy information system?
The communications that relate to energy but are generated independently from the
energy information system are very important. Firstly, they may provide a
development opportunity for you, and secondly, they may be in conflict with what you
want to do.
Next we investigate the cost of operating the energy information system. There is
little point in spending more than we save!
Audit of energy information system running costs
Item Annual Cost

(A)ctual or (E)stimated
Meter reading 825 (E)
Meter calibration Nil
Software and software support 100 (A)
Hardware and hardware support 145(A)
System operators 600(E)
External contracts Nil
TOTAL COST 1,670/year
Actual costs have been taken from records. The estimated costs have been based
on labour rates and an estimate for time spent. Is it right that there are no meter
calibration costs? Are the hardware support costs exclusively for the energy
information system?
When the collection of data and information is completed, look at the options for data
analysis that are available to you.
ARCHIVED DOCUMENT
27
Selecting Analysis Techniques

It is not possible to propose standard data analysis techniques as requirements and
possibilities will vary from system to system. There are basically two activities to be
undertaken:
1. Check existing analysis techniques to see if they are being used accurately and
appropriately. (In our example one site claimed to be using CUSUM but
investigation showed that in practice they were using moving annual totals. The
output was useful, but misleading!)
2. Take samples of raw data and apply different analytical techniques to them. The
techniques may be more, or less, sophisticated. The reason for this is simple:
The most common fault with an energy information system is not making
effective use of existing data.
There is little point in spending time and money on collecting additional data if the
prime objective - supporting energy management - can be carried out with the
existing data.
Assessing Existing Systems

You should now have a good understanding of the information sources available
and will be able to make an assessment of the existing information system. Before
drawing any conclusions, you need to assess the quality performance of the system
particularly accuracy and reliability.
Accuracy
Select, at random, an energy management report.
Check any calculations that have been made using the data in the report. For
example, if there is an SER, tonnage and energy, check that the SER is correct. If
you find problems, are these as a result of rounding errors, misunderstanding or
software errors?
When the report has been passed as correct, on the basis of its own data, you then
need to follow an audit trail back to the origins of the data. To do this, select a
number of data items, for example, energy consumption, degree days, production
volume, hours of operation. Then, taking each in turn, check them back to their
original data source e.g. the meter, log sheet, recorder.
You may want to go further and verify the data sources themselves. For
example: When was the steam meter last calibrated? Is the measure of
floor area gross or net?
Using methods similar to quality control techniques, you need to make an

assessment when errors are found - are they single errors or indicative of
system errors?
Any organisation working towards ISO 14001, Environmental Management Systems

should be prepared for the certifying body to carry out an audit process as above.
Reliability
Having checked accuracy, the next task is to look at the delivery of information. Is
it on time and in the required form? If the system is based on regular reporting,
identify all the regular reports that are promised by the system and then look at their
delivery.
Regular reporting - At a simple level, this might be 12 monthly reports and

one annual report. In this case the check would be whether all 13 reports
were produced. If the right number of reports have been produced, how
long after the month end did they appear? How long after the year end did
the annual report appear? Having established the performance, is it
acceptable? A monthly report that is eight weeks after the end of the month
has far less value than one appearing during the week after the month end.
(If you want to use a ratio to rate this area of performance you can divide
the actual time for report delivery by the target time for report delivery.
Anything less than unity is good, anything over unity indicates a missed
target.)
ARCHIVED DOCUMENT
28
Exception Reporting - It is essential that exception reporting is reliable, as

without a report the user assumes all is well. This audit has two parts. The
first is to apply the timeliness criteria from above, i.e. how quickly were the
reports provided? The second is more complex and is to check that all
exceptions that should have been reported were reported. To do this, you
will need to look at the historical data and identify the points of exception.
Were reports raised to cover these? If not, why not? Clearly if your system
does not identify exceptions then you cannot report them!
Communication
Having satisfied ourselves about quality, we need to assess the effectiveness of
communication. If reports are accurate and reliable, what happens to them? Are
they understandable? Are they useful? Try the following process to check the
effectiveness of your communication methods:
Select, at random, a number of reports that have been issued and then
arrange meetings with the recipients:
Prepare a structure for the meeting so that you can find out what recipients
do with the reports:
- Do they file, distribute or bin them!
- Do they understand the reports?
- Are reports sent to the right people?
- Can they suggest improvements?
- What actions have been taken as a result of the reports?
The output of this process will be more qualitative than quantitative. Depending on
the results of these interviews, you may decide to send a questionnaire to a wider
group or use some other means to widen the consultation.
Comparative ratios
Determining some key ratios can assist in assessing the system. The first of these
is the ratio of total energy cost and number of meters.
Taking information from delegates attending Energy Efficiency Best Practice

Programme workshops around the UK, the spread of results for energy spend per
meter has ranged from less than 2,000 to over 1 million. The 2,000/meter site
had a requirement for making all departments cost accountable. The
1 million/meter site was a major chemical plant. This Guide does not recommend
an appropriate level for energy spend per meter. The aim is to make you aware and
encourage you to assess your own situation.
Energy spend per meter
Total energy spend 1,020k/year
Total main meters 4
Total sub-meters 7
Total meters 11
Annual energy spend k/Meter 93k/year
We can now consider the results for our example company (see tables). With an
energy spend of 93,000/meter we may feel that additional investment in metering
may be worthwhile.
ARCHIVED DOCUMENT
29
When we see that the energy information system operating cost is less than 0.2%
of the total energy spend, and the total savings potential is 7%, we can see there is
good justification for developing the system further - if that will lead to additional
energy savings.
Information system ratios

Operating cost/meter 152/year
Operating cost/total energy spend 0.2%
Operating cost/savings potential 2.3%
Savings potential/total energy spend 7.0%
Based upon the following data:
Operating cost 1,670/year
Number of meters 11
Total energy spend 1,020k/year
Savings potential 71.6k/year
Taking Action
This assessment will highlight the differences between the existing information
system and the requirements you have identified for its improvement. Having
identified these areas for improvement, you can plan and implement appropriate
actions.
System development should not be carried out in isolation. For example, if a new
financial management system is being introduced, it may be an ideal opportunity to
integrate some elements of the energy information system. For a high street based
organisation it may be worth looking at how any rolling refurbishment programmes
can help.
From a clear understanding of your current position and your planned destination,
all that remains is to get there. However, there are too many examples of energy
information systems where one person takes all responsibility and attempts to carry
out all the work.
Taking action needs the support, commitment and participation of other

individuals who have a stake in the successful operation of the companys
energy management activities.
ARCHIVED DOCUMENT
30
TOOLKIT
TOOLKIT
The forms provided in the section can be copied and used as a toolkit. It is more
7
likely that you will want to customise them for your own organisation on a
spreadsheet (where you can build in calculations and do further analysis) or a word
processor.
If you work within a multi-site organisation, you may want to apply the tools to each
of the sites. Alternatively, you may wish to audit a number of sample sites. The
balance here will be between the effort and cost required and the potential gains.
Action Plan
The Action Plan has been prepared to set targets and responsibilities for an initial
review of the companys energy information system needs.
Action Plan
SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Item Responsibility Start date Completion date
System definition
Information
gathering
Analysis
Assessment
Implement
Next audit
Date Prepared: ____________________________
It would be reasonable to plan for completion of the system definition, data

gathering, analysis and assessment stages by week ten. Implementation would
then follow, with the next audit taking place 12 months after this initial review.
ARCHIVED DOCUMENT
31
Potential Savings Assessment

The starting point is to look at the annual expenditure and savings potential for each
utility used. Savings potential is your estimate of the percentage savings you think
can be made, without significant capital investment. For a site that has not
addressed the management issues of energy and workforce awareness, savings of
around 5% are normally possible.
Utility Savings Potential

SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Utility Annual % of % savings Savings

spend total spend potential (annual spend
k x % potential
savings)
Electricity
Gas
Water
Oil
Total 100%
Total savings/annual spend %
If you are running a cost-optimised energy management programme you should be

focusing your effort on the largest savings potential. If you are driven by
environmental objectives, you could prepare an alternative analysis based on
environmental impact. However, since, to an extent, cost is proportional to
environmental impact this may not be necessary.
Metered Supply - Main Meters

This table shows the meters used by the utilities to compile your invoices. On-line
metering indicates that the meters are read remotely, i.e. the utility does not have to
visit the site to read the meter. Code 5 is an example of on-line metering.
Main Metering
SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Utility Number ID/ Type Reading frequency

of meters location A = On-line metering Q = quarterly; M = monthly
B = (Basic) Manually read; W = weekly; D = daily
Other - specify 30 = 30 minutes
Other - specify
By By
utility organisation
Electricity
Gas
Water
Total
ARCHIVED DOCUMENT
32
TOOLKIT
This table covers the main utility meters. Although you may have paid for the
installation of these meters, they normally remain the property of the utility or meter
operator. If you want to connect reading devices to these meters, they will need to
be approved by the meter owner/operator.
You will find recording the meter identification and location useful if you decide to
change utility suppliers.
If you have no other meters, these meters can provide a cost effective data source
for an energy information system.
Bulk Supply - Unmetered

Include in this table each of the utilities that are supplied in bulk, e.g. oil, coal, and
perhaps water.
Unmetered Bulk Supply

SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Utility Method Frequency Consumption

of charging of charging metering
Oil
Coal
Water
Since these supplies are not metered you will be provided with supply information
rather than consumption data.
If water is supplied at a flat charge regardless of consumption, there is little financial

incentive to reduce consumption. However, changing to a metered supply may be
an option.
Invoice Tracking
First identify how many invoices are processed and their key characteristics. For
example, estimated readings will lead to problems with consumption measurement.
Direct debit will change how the Accounts Department handles the invoice, whilst
electronic invoices should be easier to enter into a system.
Invoice Tracking
SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Detail Gas Electricity Water Total
No. of monthly invoices
Invoices per year
Paper or electronic? P/E
Estimated readings? Y/N
Direct debit payment? Y/N
ARCHIVED DOCUMENT
33
You will find it useful to draw a flow diagram of the invoicing process:
Day 1
Invoice arrives
Incoming accounts
Day 1
Logged in
Day 2
Copy to engineer
Day 6
Scheduled for payment
Filed in accounts
Day 14
Paid direct debit
Flow diagram for energy invoices
Sub-meters
This table covers the meters used to break down consumption by service or
department. They may be used simply for monitoring or for departmental charging.
Sub-metering
SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Metered Number Type Reading frequency

utility of sub-meters A = on-line metering Q = quarterly; M = monthly
A = on-line, X = not read
Total
You will also want to record the locations of any sub-metering and the processes
they monitor
ARCHIVED DOCUMENT
34
TOOLKIT
Energy Drivers
The energy consumption of your site is driven by a number of factors, such as the
weather, tonnes of product, moisture content of raw material and hours worked.
For this Section you need to list what you think are the significant energy drivers
for your site.
Energy Drivers
SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Driver Process Source of data
Communications
This table is used to plot energy reporting that takes place within the
site/organisation - regardless of origination (i.e. not just what is produced by energy
management).
Energy Reporting
SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Report Frequency From To
Other Relevant Communicated Methods
Type To From
ARCHIVED DOCUMENT
35
Running Costs
This looks at the costs of operating the energy information system. This is done
using annual costs with best estimates where actual figures are not known.
Complete the first column in year one - then compare this to the planned or actual
costs for year two.
Energy System Operating Costs

SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Item Annual cost year 1 Annual cost year 2

(A)ctual or (E)stimated (A)ctual or (E)stimated
Meter reading
Meter calibration
Software and
software support
Hardware and
hardware support
System operators
External contacts
Total cost
Notes:
Meter reading - This is the staff cost associated with reading the meters.
It may be a dedicated job or a small part of another function.
Meter calibration - This is normally required for steam meters, although

other meters may require re-calibration
Software and software support - You may have to pay annual fees for
software, and/or its maintenance.
Hardware and hardware support - This will be for any dedicated

hardware, or any hardware costs that are charged to the system
System operators - This covers the staff cost of the energy information
system operators.
ARCHIVED DOCUMENT
36
TOOLKIT
System Performance Ratios
Energy Spend Per Meter

SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Total energy spend /year
Total main meters
Total sub-meters
Total meters
Annual spend/meter /year
You would expect the energy spend per meter to fall year on year as a result of
energy efficiency measures. However, in the early stages it may fall as a result of
increased awareness through better metering.
Information System Ratios

SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Operating cost/meter k/year
Operating cost/total energy spend %
Operating cost/savings potential %
Savings potential/total energy spend %
Based upon the following data:
Operating cost k/year
Number of meters
Total energy spend k/year
Savings potential k/year
As guidance, you should expect operating cost/total energy spend to be a fraction

of the potential savings/total energy spend. If it is not, you are spending more than
you are saving! You also need to be comfortable with the operating cost per meter.
ARCHIVED DOCUMENT
37
Progress
To review progress over a series of annual audits you might like to use the table
below.
Progress Record
SITE DETAILS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Audit 1 Audit 2 Audit 3
Date
Annual energy spend k
Savings potential %
Total savings/total spend %
No. of main meters
No. of sub meters
System operating cost k
Spend/meter k
Operating cost/meter k
Energy savings achieved k
Most of the figures in this table can be taken from the tools in this section. The
exception is the last row Energy savings achieved, which must be added as
appropriate.
Comparing performance against previous audits enables you to identify progress in

energy savings or enhancing the energy information system. It may be useful to
compare trends in some of the key indicators too.
ARCHIVED DOCUMENT
38
The Governments Energy Efficiency Best Practice Programme provides impartial, Energy Consumption Guides: compare energy use in
authoritative information on energy efficiency techniques and technologies in industry, transport specific processes, operations, plant and building types.
and buildings. This information is disseminated through publications, videos and software,
Good Practice: promotes proven energy efficient techniques
together with seminars, workshops and other events. Publications within the Best Practice through Guides and Case Studies.
Programme are shown opposite.
New Practice: monitors first commercial applications of new
energy efficiency measures.
Further information
Future Practice: reports on joint R & D ventures into new
energy efficiency measures.
For buildings-related publications For industrial and transport publications
please contact: please contact: General Information: describes concepts and approaches
Enquiries Bureau Energy Efficiency Enquiries Bureau yet to be fully established as good practice.
BRECSU ETSU
Fuel Efficiency Booklets: give detailed information on
Building Research Establishment Harwell, Didcot, Oxfordshire, specific technologies and techniques.
Garston, Watford, WD2 7JR OX11 0RA
Tel 01923 664258 Fax 01235 433066 Energy Efficiency in Buildings: helps new energy managers
Fax 01923 664787 Helpline Tel 0800 585794 understand the use and costs of heating, lighting etc.
ARCHIVED DOCUMENT
E-mail [email protected] Helpline E-mail [email protected]
CROWN COPYRIGHT FIRST PRINTED MARCH 1998

GPG231 Introducing Information Systems For Energy Management

Uploaded by

Copyright:

Available Formats

GPG231 Introducing Information Systems For Energy Management

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

GPG231 Introducing Information Systems For Energy Management

Uploaded by

Copyright:

Available Formats

GOOD PRACTICE GUIDE 231

Introducing information systems for

ARCHIVED DOCUMENT BEST PRACTICE

The John Pooley Consultancy

LIST OF RELEVANT GOOD PRACTICE GUIDES

111. MONITORING AND TARGETING IN FOUNDRIES

Copies of these Guides may be obtained from:

Energy Efficiency Enquiries Bureau

NORTHERN IRELAND SCOTLAND WALES

4 DATA ANALYSIS TECHNIQUES 12

6 ASSESSING ENERGY INFORMATION SYSTEM 23

Energy Information and Energy Management

An energy information system may be stand-alone, part of an integrated system or

For an engineering plant, an energy information system might comprise meters

However, energy information systems are also needed to support environmental

It is less than 15 years since the PC was introduced.

Most energy managers now have easy access to a reasonable

There have been significant cost and performance improvements in

Many organisations have access to electronic mail and the Internet.

How to Use This Guide

The Operational Cycle

Data analysis Action

The operational cycle

Data, Information and Knowledge

a listing of energy meter readings;

a record of maximum and minimum external temperatures;

Energy information systems

Whilst the prime purpose of an energy information system is to support energy

For an individual process it may be possible to use a simple graph or chart

For a single site, a spreadsheet developed in-house might be appropriate,

For a site with a large number of meters, the solution may be a

For a complex, multi-site operation the solution may be a bureau-based

Systems may be further sophisticated with an expert- or knowledge-based system

Visual reading Single PC based On-line metering,

The range of solutions

At one time, information systems were limited by a lack of low-cost, effective

GPCS 330, A Comprehensive Energy Management Information System at

What is the purpose of the system?

Data Types and Sources

There are two main types:

Activity drivers. An activity driver is a feature of the organisations activity

Condition drivers. A condition driver is one where the influence is not

Meters and Metering Periods

Meters and sub-meters

A feature of sub-metering which must be appreciated is that for a totally sub-

More meters and more frequent reading?

A month may be 28, 29, 30 or 31 days, or even 35 days when a five-week

it has a universal definition - 7 days;

if a reading is missed, it is only one of 52, rather than one of 12;

suitable sets for data analysis can be generated quicker;

The major disadvantage of weekly reporting is the increased amount of data

Data Collection Methods

number of data collection points;

method of data collection;

frequency of data collection.

Physical distance is also relevant. Single sites such as an international airport, a

Collecting data from invoices

confirmation of site and tariff details;

comparison of utility reading with organisations own reading.