Foreword

We live in a changing world. Not only this, the pace of change is itself chang-
ing. We have entered an era of incredible universal transition that will influence
various aspects of our lives. Today, the energy and technological worlds are
witnessing drastic changes that are influenced by intertwined causes of growth:
energy demand, rapid transition of energy systems, political factors, technologi-
cal evolution, and societal and cultural implications.
These changes simultaneously depend on and influence the way we live,
the way we work, and the way we move. There is nothing more crucial than
bringing together leadership power, domain expertise, knowledge, and the many
lessons of past civilizations and industrial revolutions.
In the energy sector, the petroleum industry is still the most important source
of energy and the provider of primary material to many industries. This industry
is also influenced by the transformative technologies that offer tremendous po-
tential to redraw not only the physical infrastructure but also the intellectual cli-
mate, business models, and asset management systems. This petroleum industry
is very different from other industries. It is among the few industries that are
perceived as generators of big data through the various phases from exploration
to abandonment. It is an industry that has been treasuring all sorts of data for
decades as it relies heavily on these data in managing its assets and running its
operations. Companies in the petroleum industry are risk averse and have strict
safety standards due to the nature of their business. None of these companies
can afford to operate without appropriate decision-making processes that rely
greatly on adequate technologies, experience, and data analytics. The recent
advancements in the digital world including the increase in computer power
and information processing, the speed of communication, and the reduction in
cost of data storage; most of the companies in the petroleum industry expedited
and extended their digital business with great ambitions to extract or generate
new value and identify potential opportunities from both their old and new data
augmented with either existing or new business models.
Many of these opportunities can be identified if the underlying mechanism
is exposed in numerical format and encapsulated in the language of mathemati-
cal formulas, the language of data science. It is the purpose of data science to
obtain, clean, and curate the data necessary and the purpose of machine learn-
ing to produce the formulas (i.e., new knowledge). With this new knowledge,
managers of assets are expected to do better.

xiii
xiv Foreword

The authors of this book have stepped back from debating what might be
achievable and instead they started with the more fundamental exploration of
the potential of machine learning and demonstrated how it can help the petro-
leum industry. They have clearly indicated that machine learning in itself is not
an objective but a means. No one is expected to implement a technology for the
sake of it or because of hype around it, but for the benefits it may bring.
Today, machine learning is among the most widely used data analytics tech-
niques in various industries including health care, education, manufacturing,
automotive, retail, finance, logistics, transportation, and energy including oil
and gas. No one denies the value that this technology brings if applied appro-
priately and at large scale. I can see that there will eventually be no industry
that will be left untouched by machine learning, as it possesses the potential to
empower humans to do what may be perceived today as impossible. In the oil
and gas industry, machine learning has already been implemented or tested in a
wide range of business sectors and the industry is very ambitious about extend-
ing its applications and very optimistic about its positive impact.
Although many of the challenges in the petroleum industry are global, many
solutions, assessment studies, and remedial actions are almost local. Only few
companies are effective at transferring and exchanging experience and learning
from their successes and failures. This book is a clear testimony of the desire of
its authors to demonstrate the meaning and importance of sharing local knowl-
edge globally.
This book presents an overview of the field of machine learning and points
toward the many used cases in the oil and gas industry where these have al-
ready solved problems. The hype surrounding machine learning will hopefully
be cleared up as this book focuses on what can realistically be done with ex-
isting tools. It aims to present much of the fundamental understanding of the
petroleum industry and machine learning in a way that is comfortable for all
to follow. It summarizes the fundamental concepts of various applications of
machine learning in the oil and gas industry and does not pretend to present a
comprehensive review of the details though references to books providing such
reviews are included where appropriate. While this book will not make you into
a machine learner, it will provide enough knowledge to conduct yourself ef-
fectively with data scientists and also to be able to reflect on the overall quality
and results of a project.
The book addresses four main groups of readers: Oil and gas professionals,
machine learners and data scientists, students and researchers, and the general
public.
The primary audience is any person working in the oil and gas industry who
wants to understand what machine learning is and how it applies to the industry.
Machine learners and data scientists will learn about the oil and gas industry
and its complexities as well as the use cases that their methods can be put to
in this industry. They will learn what an oil and gas professional expects to see
from the technology and the final outcome. The book puts into perspective some
 Foreword xv

of the issues that take center stage for data scientists, such as training time and
model accuracy, and relativizes these to the needs of the end user.
This book also targets graduate students and researchers at the cutting edge
of investigations into the fundamentals and applications of machine learning in
the oil and gas industry.
For the general public, this book presents an overview of the state-of-the-art
in applying a hyped field like machine learning to an old-world industry.
The book can be divided into two parts. The chapters of the first part discuss
the petroleum industry and machine learning with associated management chal-
lenges. The second half focuses on practical case studies that have been carried
out by oil companies and report on what has been done already as well as what
the field is capable of. In this context, the reader will be able to judge how much
of the marketing surrounding machine learning is hype and how much is reality.
This book will have served its purpose if an oil and gas company uses the
advice given here in facilitating a machine learning project or toolset to drive
value. It is meant partially as an instruction manual and partially as an inspira-
tion for oil company managers who want to use machine learning and artificial
intelligence to improve the industry and its efficiency.
As such, this book is deliberately designed to be a passionate discussion
starter for what I consider to be one of the digital world’s most important de-
bates. I believe the authors were successful in their mission to open up and
stimulate further discussions. In a rapid transition of energy systems, such de-
bates are essential and can only be achieved by fundamental research, free from
economic and political constraints, and by people free to rethink our current
challenging business environment.
I truly believe that a main reasons for societies and industries to continue,
grow, and flourish is the existence of distinguished members who generously
share their valuable experience, knowledge, and inspiration.
I am deeply grateful to the leading author, Patrick Bangert, and the other
authors for writing this book. Moreover, I hope that this book inspires you to
think deeply about the opportunities machine learning can create, and I do invite
you to contribute to this journey by sharing your story.

Saeed M. Al Mubarak
Chairman, SPE Digital Energy Technical Section (2017–23)
Chairman of the Standards Committee and Chairman of
the P&FDD Academy Technical Program, Saudi Aramco