We spoke to Louis Anthony Cox, Jr. about the ideas and inspiration behind his recently published book, Breakthroughs in Decision Science and Risk Analysis.
Focusing on modern advances and innovations in the field of decision analysis (DA), Breakthroughs in Decision Science and Risk Analysis presents theories and methods for making, improving, and learning from significant practical decisions. The book explains these new methods and important applications in an accessible and stimulating style for readers from multiple backgrounds, including psychology, economics, statistics, engineering, risk analysis, operations research, and management science.
Highlighting topics not conventionally found in DA textbooks, the book illustrates genuine advances in practical decision science, including developments and trends that depart from, or break with, the standard axiomatic DA paradigm in fundamental and useful ways. The book features methods for coping with realistic decision-making challenges such as online adaptive learning algorithms, innovations in robust decision-making, and the use of a variety of models to explain available data and recommend actions. In addition, the book illustrates how these techniques can be applied to dramatically improve risk management decisions.
Louis Anthony (Tony) Cox, Jr. PhD is Chief Sciences Officer of NextHealth Technologies, a Denver-based health care advanced analytics software company and President of Cox Associates, Inc., a Denver-based applied research company specializing in quantitative health risk assessment, risk analysis, causal modeling, and operations research. Professor Cox is also Clinical Professor of Biostatistics and Informatics at the University of Colorado at Denver, a member of the National Academy of Engineering, and Editor-in-Chief of Risk Analysis: An International Journal.
1. Congratulations to you on the upcoming publication of your book, Breakthroughs in Decision Science and Risk Analysis, which is also the first book to be published in the Wiley Essentials in Operations Research and Management Science book series. How did the writing process begin?
Thank you. This book was largely inspired by the challenge of communicating to a wide business and analytics audience some of the key breakthroughs mentioned in the Wiley Encyclopedia of Operations Research and Management Science (EORMS). The Encyclopedia is an eight-volume work that presents the state-of-the-art of analytics and the OR/MS field through clear, up-to-date, authoritative articles. A key challenge is how to distill and extend some of the key advances to reach a wider audience that does not necessarily specialize in analytics. The Editor-in-Chief, Jim Cochran, and I realized that EORMS is a terrific resource for operations research and analytics specialists, but there are really important and valuable new methods and idea for assessing risks and making better decisions that many other people in business and policy circles could also benefit from. These deserve to be pulled together and presented in a highly accessible way and in greater detail than the encyclopedia article format. So Wiley, Jim, and I decided to create a book in which assembled my “dream team” of top authors to contribute the most important things they had to say about practical advances in decision science and risk analysis to a wider audience of decision and policy makers, as well as analytics professionals.
The main idea was to identify and explain genuine breakthroughs that make it possible for people to make decisions under risk and uncertainty better now than was possible a few decades ago. I outlined key chapters and themes, such as breakthroughs in adaptive learning methods, simulation-optimization, new insights from behavioral economics and decision psychology of risk, and so forth. Then, I sent invitations to authors whom I regard as both thought leaders in these areas and also great expositors, who know how to explain even sophisticated concepts so that they will be useful to practitioners. We discussed and improved the initial outline and early drafts of the chapters, and then created a final book that I think will really help practitioners in fields as diverse as healthcare, health and safety regulation, manufacturing, and Homeland Security understand what the new methods have to offer for improving how we make decisions and assess and manage risks.
2. Who should read the book and why?
This book is really directed toward practitioners who want to know about new ideas they can use to make better decisions about what to do when the consequences of different choices are uncertain. There is a very rich legacy of great ideas from the 20th century, such as expected utility theory and the more psychologically realistic Prospect Theory, that have formed the core of decision science for many decades. This book briefly summarizes these ideas, but its focus is on what else is available, what’s new and useful, and how to help make decisions when those classical ideas are difficult or impossible to apply — for example, when there is no good basis for assessing credible probabilities, or when utilities are unknown, or when we simply need to make decisions that are more defensible and robust to uncertainties than those older methods. The book is meant for people who want to know what is new and useful in practice, and who want an accessible, authoritative explanations written by top experts but designed for practitioners who may not be specialists in analytics.
3. What is it about the area of risk analysis that fascinates you?
Everything about risk analysis fascinates me. I think it’s mainly because so much of what we do in life that is important has very uncertain consequences, and seeing how far analysis and reason can take us in making effective decisions in such situations is both intellectually challenging and of tremendous practical value. That there are useful frameworks for thinking about how to behave despite great uncertainties is a very encouraging advance that is equally fascinating from the technical, psychological, and philosophical angles. Plus, progress in decision-making under uncertainty and risk has many practical applications, to everything from combating terrorism to making healthcare systems run better, and these applications are fascinating and important in their own right. The very idea of risk is fairly modern and technically interesting, and the fact that it can be usefully quantified and analyzed in many situations is a breakthrough that can transform how we understand and manage risks in practice.
“So much of what we do in life that is important has very uncertain consequences, and seeing how far analysis and reason can take us in making effective decisions is both intellectually challenging and of tremendous practical value.”
4. Why is this book of particular interest now?
The book is timely because we live in a society where interconnected systems, networks, infrastructures, and economies are creating new types and increasing scales of risks, while public perceptions of risk are under a constant barrage of manipulation from politicians, researchers, activists, reporters and media, and others who understand that changing voter and consumer preferences and behaviors is often more easily accomplished by engaging concerns about risks than by presenting compelling logical and factual arguments and analyses. Recognizing and overcoming such manipulations and responding more effectively to the very real and increasing challenges of managing risks in increasingly complex and tightly interconnected business, technology, financial, social, and economic systems requires breakthroughs, and that is what this book is about.
Risks from intelligent adversaries, cyber criminals, systemic risks in financial systems, decaying infrastructures and the like have also heightened public awareness of the need for highly intelligent and effective countermeasures. This demand for more effective, rational, and responsible risk management has been matched by a surge in the supply of innovative, creative, and practical ideas for how to manage risks better. This book aims to bring the two together, showing how new ideas can be applied to new as well as old risk management problems.
5. What were your main objectives during the writing process? What did you set out to achieve in reaching your readers?
The compelling vision behind this book was the need to explain genuine breakthroughs in decision science and risk analysis that can really help us to manage important risks better. I felt that someone needed to explain these powerful new methods – concepts and techniques that did not even exist a few decades ago, and that demonstrably improve on older ideas in practical applications. Explaining those in a way that most people can understand using simple, clear, yet authoritative expositions, is the main purpose of this book.
6. Were there areas of the book that you found particularly challenging to write, and if so, why?
There have been some remarkable accomplishments in artificial intelligence and machine learning in the last ten years, and a lot of that information is not very visible to people trained in decision analysis. I wrote some chapters to try to bridge the gap between those two communities, but I think that integrating concepts such as low-regret reinforcement learning algorithms from machine learning, developed mainly by computer scientists, with decision science and risk analysis concepts developed largely by business and engineering communities and psychologists, requires quite a lot of flexibility from both authors and readers — a willingness to pick up useful ideas wherever they may be found, and to apply them wherever they are most useful.
“The compelling vision behind this book was the need to explain genuine breakthroughs in decision science and risk analysis that can really help us to manage important risks better.”
7. Please could you tell us more about your educational background and what it was that brought you to decision science and risk analysis in the first place?
I am the world’s first Ph.D. in Risk Analysis (MIT, 1986). I have been fascinated by rational decision-making and risk assessment and management ever since I read Duncan Luce’s and Howard Raiffa’s wonderful book Games and Decisions in high school. I was later privileged to study with Luce at Harvard, and to learn engineering and operations research approaches to risk analysis and decision analysis at MIT. At the same time, I was working for a wonderful management and engineering consulting company, Arthur D. Little, where I constantly ran into new problems that required going beyond existing textbook models and methods. That combination of studying great ideas with great teachers while working on real-world problems that demanded more inspired in me both a deep admiration and appreciation of what existing decision science and risk analysis methods could do, and also an awareness of what limitations that had that needed to be overcome in practice. Breakthroughs in Decision Science and Risk Analysis is a big step in that direction.
8. You’re also an Area Editor on the Wiley Encyclopedia of Operations Research and Management Science. What was it like working on that project?
Well, working on the encyclopedia has been thrilling. It gives me a unique opportunity to work with some of the best minds in the world on defining and explaining state-of-the-art advanced analytics and its many applications in operations research and management science.
9. How do you manage to juggle editing Risk Analysis: An International Journal and your many other commitments?
I find that editing the journal actually makes it easier to do other fun things. The journal has such a wealth of exciting ideas and applications that other projects can draw and build upon that I find it enriches, and makes easier to think about, everything else I do.
9. Who should read the journal and why?
Risk Analysis: An International Journal is the official journal of the Society for Risk Analysis, and is read by its members. But it also has a much wider readership of analysts and policy makers involved in risk regulation, risk psychology, and applied risk analysis. The journal’s main focus is on public health, safety, and environmental (HS&E) risk analysis, including risk assessment, risk perception, risk communication, and risk management, with emphasis on important real-world applications where analysis demonstrably improves the management of HS&E risks.
The journal has areas devoted to Policy, Mathematical Modeling, Microbial Risk Assessment, Engineering, Risk Perception, Risk Communication, Decision Sciences, Health Risk Assessment, and Environmental and Ecological Risk Analysis. We have an extraordinarily talented team of Area Editors for these areas, all of whom are thought leaders in their respective fields, and they have helped to make the journal well worth reading for anyone with interests in one or more of these areas. The journal is truly interdisciplinary, and helps set the standard for important advances in risk analysis methods and applications.
“Integrating concepts requires quite a lot of flexibility from both authors and readers — a willingness to pick up useful ideas wherever they may be found, and to apply them wherever they are most useful.”
10. Are there any hot areas that you’d like to see the journal publish in? The journal published a special issue on the Economics of Risk Analysis in spring 2014. Are there any other special issues in the pipeline?
Yes, we have quite a few other special issues on hot topics in the works. We have special issues in the pipeline now consisting of new papers on Risk, Perception, and Response; Infrastructure Risk Assessment; Chemical Alternatives Assessment; Validating Adversarial Risk Models; Games, Decisions, and Reliability; and Research Synthesis, among others. Wiley, the Society for Risk Analysis and Risk Analysis: An International Journal also make some virtual online issues available for free to the public via several web sites, including that of the Society for Risk Analysis (http://www.sra.org/sra-journal). These consist of some of the journal’s greatest-hit papers on specific topics, in the works. Topics covered in forthcoming virtual issues include Foundations of Risk Analysis, Top Ten Accomplishments in Risk Analysis, and possibly Bayesian Networks and Risk Matrices.
11. Are there people or events that have been influential to your career?
Yes, I have benefited greatly from some wonderful teachers, colleagues, and friends throughout my career. My Ph.D. advisor, Professor Alvin W. Drake of MIT (now deceased) was an inspiration. He taught me to think more clearly about applied probability and decision analysis. He was also a funny, kind, warm, and deeply insightful friend who first suggested that I should seek a Ph.D. in Risk Analysis, based on the passion he saw in me for that nascent field. Duncan Luce and Art Dempster of Harvard taught me to understand mathematical modeling and statistical analysis in ways that might be characterized as fundamental — focusing on what models and their assumptions really mean and whether they make sense first, and on techniques for computing answers second.
Two past Editors-in-Chief of Risk Analysis, Curtis Travis and Mike Greenberg, helped draw me more deeply into the field, and into thinking and writing about it. Curtis heard a presentation that I gave in the early eighties at my first professional conference, based on work I was doing at the time for the Electric Power Research Institute. He came up to me afterwards and invited me to write up the results for the new journal Risk Analysis. I have been contributing articles ever since. Decades later, Mike Greenberg, the Editor of Risk Analysis before me, repeatedly encouraged me to write about interesting topics and to find ways to bring win-win discussions and insights out of discussions where authors have sharply differing (and sometimes sharply worded) views.
Mike and I collaborated on a number of new ideas for the journal, including special and virtual issues, and his insights and friendship have helped make working on the journal exciting and fun for much of the last decade. I have also been blessed with many friends and colleagues who share my enthusiasm for risk analysis. Several work with me now on the journal. Others, such as Professor Jerry Brown at the Naval Postgraduate School and Professor Vicki Bier at the University of Wisconsin-Madison (who contributed two chapters to Breakthroughs) have been friends, collaborators, and co-authors for many years, and have pushed me to think more clearly about topics, such as terrorism risk analysis, that we all care about.
I feel I have been richly blessed with people who have helped to shape my career and who have improved my own understanding of risk analysis. There is still a great deal to learn and to make useful about decision science and risk analysis, but also a great deal has been accomplished, and I am grateful to Jim Cochran and the Wiley staff of EORMS for making possible Breakthroughs in Decision Science and Risk Analysis to explain some of the most exciting and worthwhile advances that are making a real difference — and deserve to make a larger one — in changing risk management of our complex and uncertain world for the better.