About the Authors

Lev Virine has more than 20 years of experience as a structural engineer, software developer, and project manager. In past 10 years he has been involved in a number of major projects performed by Fortune 500 companies and government agencies to establish effective decision analysis and risk management processes, as well as to conduct risk analyses of complex projects. He also has experience in the information technology, manufacturing, pharmaceutical, and oil and gas industries.

Lev is the author of more than 25 scientific papers and patents. His current research interests include decision analysis and risk management, particularly their applications in project management. He writes and speaks to conferences around the world on project decision analysis, including the psychology of judgment and decision-making, modeling of business processes, and risk management. He received his doctoral degree in engineering and computer science from Moscow State University. Lev can be reached at lvirine@projectdecisions.org.

Michael Trumper has worked in the fields of technical communications, marketing, and software development for the past 16 years. Over the past eight years, he has been involved in projects involving economic valuation and risk and project lifecycle modeling. Michael has authored several papers and articles on project risk management. He holds a bachelor’s degree from the University of Victoria and is currently a master’s candidate in the University of Liverpool’s laureate program. Michael can be reached at mtrumper@projectdecisions.org.

Lev and Michael are located in Calgary, Alberta, Canada. For supplemental information about their work on project decision analysis and risk management, visit www.projectdecisions.org.

CHAPTER 1

Project Decision Analysis: What Is It?

Most of us believe we are pretty good at making decisions, yet we continue to make poor ones. And over time our poor decisions become a burden that we impose on each other, especially when the decisions we make as managers are connected to large-scale projects that affect many people. The process known as structured decision analysis—which is described in detail in this book—can improve our ability to make better decisions, particularly in project management, where the decisions can be complex. Indeed, today many organizations in both the public and private sectors use decision analysis to solve their project management problems.

THE BURDEN OF POOR DECISION-MAKING

In 2004—2005, Governor Arnold Schwarzenegger of California was involved in a complex decision-making process. He was not considering a role for his next action movie, nor was he selecting a new energy weapon to blast villains in a sci-fi movie. This was something more serious: the governor involved himself in the design process for a new bridge in San Francisco (Cabanatuan 2005).

This was not just any bridge. The $6.3 billion project (Figure 1.1) was to replace the existing Bay Bridge. The original plans called for the section of the bridge east of Yerba Buena Island to include a huge suspension span. Although the construction of the foundations for the suspension span had started a few years earlier, the governor’s office insisted that a simple viaduct would be cheaper and faster to build. Transportation officials did not agree, believing that a design change from a suspension span to a viaduct would slow construction.

Early in 2005 the governor’s side appeared to have prevailed: work on the foundation was halted, and the contract was terminated. A few months later, however, following a detailed analysis, both sides agreed to follow the original design, which included the suspension span. In the end, the fight over the bridge design cost $81 million to stop and restart work on the foundations for the tower that will support the suspension span. State funds and increased toll revenue (tolls were raised to $4) from state-owned bridges would cover the burden of the cost overrun from the governor’s costly decision to delay the project.

Wrong decisions are a burden that we impose on each other.

If you don’t live in northern California, you may not be directly affected by the Bay Bridge cost overrun. But, directly or indirectly, at some time you will pay for somebody’s wrong decision regardless of where you live or what you do. This is because, for example:

 Costs related to problems in developing new drugs are passed on to consumers in the form of higher prices for medications.

 Dry wells lead to increased costs for oil and gas exploration and production, leading in turn to higher prices at the gas pump.

 Governments sometimes implement ill-considered policies that can adversely affect taxes.

 You sometimes make wrong decisions yourself. The cheap brand of deck coat that you used to save a couple of dollars is already peeling off and you will have to paint your deck again next year (hopefully with a better brand).

Problems result from poor decision-making, whether the decision-maker is the manager at the pharmaceutical company, the geologist making a poor choice of where to drill for oil, the ineffective government bureaucrat or legislator making policy for the wrong reasons, or even you trying to save a few bucks by buying a low-priced deck coat.

We human beings have been making poor decisions since we first developed the ability—and the necessity—to make choices. In the modern world, however, due to the complexity and cost of projects, the price we pay for poor decisions has significantly increased. The overall cost of wrong decisions is very hard to estimate, but it is undoubtedly enormous. Say, for example, we design a multibillon-dollar oil pipeline but make a wrong decision about running it through a particular location. Then we have to move it—a step that might increase the project costs by millions of dollars. Who pays that cost? It is passed on to somebody—investors, consumers, or the government.

Poor decision-making in the medical field can have expensive, even fatal, results. The causes of medical mistakes differ. Sometimes the cause is a flaw in a hospital procedure. Most medical mistakes, however, are related to errors in human judgment.

Annually, 44,000 to 98,000 deaths occur due to medical errors (Kohn, Corrigan, and Donaldson 2000). That number would represent approximately 1.8 to 4.0 percent of the 2.4 million deaths that the Centers for Disease Control (CDC) reported in 1999. As a point of comparison for other causes of death, the CDC also reported that in 1999 there were 68,399 deaths from diabetes, 63,730 from influenza and pneumonia, and 44,536 from Alzheimer’s disease. A 2004 study (Adams 2004) put the number much higher, at 195,000 people killed each year in U.S. hospitals due to medical errors.

Similar figures on the impact of poor decision-making are available from the oil industry (Rose 2001). An exploratory well drilled some distance from an existing field is called a wildcat. A wildcat chance represents the ratio of oil and gas discoveries to wildcats (Table 1.1).

Exploratory drilling is always a risky business. Nevertheless, the wildcat chance can be improved by making better decisions: for example, by avoiding an incorrect interpretation of geological data, which was the primary source of dry holes in more than 40 percent of cases (Rose 2001). Even a slight improvement in the process of deciding where to drill wildcats would have significant economic benefits because drilling a well can cost millions of dollars.

Table 1.1 Global Discoveries (excludes U.S. and Canada)

WHY DO WE MAKE WRONG DECISIONS?

Lawrence Phillips, a prominent decision analysis expert, cites a curious paradox: Although the ability to make right decisions is considered a main indicator of project-management professionalism, many project managers are unwilling to try to improve the quality of their decisions (Goodwin and Wright 2004). Phillips suggests that many people consider decision-making to be merely an automatic process, as natural as breathing. And if we don’t need to learn how to breathe, why do we need to learn how to make better decisions? With such a blasé attitude, many project managers don’t make the effort to understand decision analysis, or they believe that it is just a theoretical discipline with no practical use in their work.

If you were asked to rate your decision-making ability, most likely you would rate yourself as “better than average.” The “better-than-average effect,” where people tend to rate themselves as above average when asked to characterize their abilities, is a common psychological bias (Massey, Robinson and Kaniel 2006) that is applicable not only to self-assessments of decision-making but also to other activities. But if we believe that we are such good decision-makers, why do we often make poor ones?

The answer resides in the fact that most of today’s important project-management decisions are complex. Without proper analysis it is hard to make choices between alternatives. Every day, project managers make numerous decisions. Most of them are trivial and do not require sophisticated analysis. If a component for your construction project is delayed, you might decide to call the supplier. Obviously, in making this choice, you can rely on common sense. You do not need to perform an advanced analysis, solve a few differential equations, or run a complex simulation model. But if you need to select a new supplier, the situation is quite different. A great deal is at stake, and a wrong decision could be very costly. Plus, there are many alternatives. Now, relying solely on your intuition may not be enough; you probably should perform a decision analysis.

Why is decision-making so complicated? There are a number of reasons:

 Most problems in project management involve multiple objectives (Goodwin and Wright 2004). An example of a project with multiple conflicting objectives is General Motors’s EV1 electric-car program. The car was introduced in 1997 to demonstrate GM’s corporate commitment to a clean environment and at the same time show the commercial viability of the technology (GM 2006). But GM pulled the plug on the project in 2002, citing insufficient public support. The automaker eventually collected and destroyed almost all of the 1,000 EV1 cars, prompting the making of a documentary titled Who Killed the Electric Car? The film was widely praised by environmentalists and others concerned about growing CO2 emissions and the country’s dependence on oil. Later, GM Chief Executive Rick Wagoner admitted that killing the $1 billion EV1 program was his worst decision: Although it did not affect GM’s profitability, it did hurt the company’s image.

 Project managers deal with uncertainties . Predicting the future is not an easy task. Selecting alternatives is the primary objective of decision analysis. Decision analysis offers tools to help project managers deal with uncertainties.

 Project management problems can be complex . The number of alternatives you face in managing a project can be significant. Decisions are usually made sequentially, based on previous decisions. But understanding how each decision will affect subsequent ones is difficult.

 Most projects include multiple stakeholders . Project managers deal with clients, project team members, project sponsors, and subcontractors, among others. All these stakeholders have different objectives and preferences.

APPROACHES TO DECISION-MAKING

In general, there are three approaches to decision-making. They are not delineated by any strict boundaries, but it is helpful to classify them to understand how we make decisions.

1. The intuitive approach . A project manager is sitting in an office (or cubicle) thinking about a decision that has to be made. After a few vacant stares out the window and several laps around the office, the manager finally feels comfortable with the alternatives and chooses one that “feels” best. The manager may not have all the information needed to make the decision, but intuitively he or she thinks that it is the right way to go. Is it a good decision? It may be. Often, however, nobody knows the truth because the results are rarely reviewed. Sometimes these types of “felt” decisions are made by groups, but that does not change their intuitive nature.

2. The advocacy-based approach . This is the most common way that decisions are made in organizations. The project manager states the problem and asks team members to perform an evaluation. But, as in the case with the intuitive approach, everything ultimately depends on the manager’s gut feeling. If the manager agrees with the evaluation, he or she will take it and make a decision. Otherwise, the manager will request that the team rework the evaluation (Skinner 1999).

3. The decision analysis approach . Under the decision analysis process, choices are made based more on the results of analysis and less on the intuition of the decision maker. The decision analysis approach entails a logical analysis of a correctly structured problem, identification of creative alternatives based on reliable information, implementation of the selected alternative, and an evaluation of the results.

DECISION ANALYSIS AS A PROCESS

Having explained what decision analysis encompasses, we must still ask: What is it? First of all, decision analysis is a tool to solve problems. It is a practical framework of methods and tools to promote creativity and help people make better decisions (Keeney 1982).

As a project manager, you don’t need to know all the details about these methods and tools, some of which can be very complicated. It is important, however, that you know two basic things that affect how decisions are made:

 We are all subject to common psychological pitfalls . People come hardwired with some psychological constructs that can mislead them when they make project decisions. If you are estimating projects costs, identifying possible risks, selecting viable alternatives, or identifying the most important project objectives, you can make predictable mental mistakes. A basic knowledge of these pitfalls and how they can affect decision-making will help you avoid them.

 We can use decision analysis techniques to avoid those pitfalls . These techniques will improve your ability to make better decisions. Moreover, most of these techniques can be applied in other areas of practice, such as financial analysis.

What you really need to know in general is that project decision analysis is a scalable and flexible process that is both practical and effective. And it is important to understand that decision analysis is not a process that creates an additional level of bureaucracy. It can be integrated into other processes that are defined in the PMBOK^® Guide (Project Management Institute 2004). We recommend that you begin to establish this process by improving your own thinking processes rather than setting it up at the organizational level.

The process includes four major phases (which are discussed in the following parts of this book):

1. Decision framing, or structuring the problem (Part 2)

2. Modeling the alternatives (Part 3)

3. Quantitative analysis (Part 4)

4. Implementation, monitoring, and review of the decisions (Part 5).

Each phase of the process involves several steps, which will be presented in our discussion of each phase.

Often, project managers believe that decision analysis is a type of cost-benefit analysis. Cost-benefit analysis is a technique used to compare the various costs associated with a project with the benefits that it supposed to return. In comparison, decision analysis is a much broader process that takes into account many parameters and uncertainties. It focuses on developing a more complete analysis of a project and understanding the ramifications of the possible choices facing a project manager.

NORMATIVE AND DESCRIPTIVE DECISION THEORY

The foundation of decision analysis is decision theory, which is the study of how to make better choices when faced with uncertainties. Normative decision theory describes how people should make decisions; descriptive decision theory describes how people actually make decisions.

To distinguish between the normative and the descriptive approaches, let’s look at decisions related to recovering a hidden treasure. The movie “National Treasure,” starring Nicolas Cage, follows a team of treasure hunters as they methodically and logically unravel a series of extremely convoluted clues. This is an example of normative decision theory because it shows how people should behave if they want to recover a treasure. Stanley Kramer’s movie It’s a Mad, Mad, Mad World (1963) is an example of descriptive decision theory, for it shows how people actually behave when they try to recover a treasure. In trying to find the treasure, instead of acting logically, the characters behave spontaneously and irrationally. Chaos and hilarity ensue, but no treasure is found.

DRIVING FORCES BEHIND PROJECT DECISION ANALYSIS

Realizing that poor decision-making in large-scale projects can result in high costs and cause harm, governments and private businesses are recognizing the importance of instituting decision analysis techniques.

The U.S. Government Performance Results Act of 1993 states that “waste and inefficiency in Federal programs undermine the confidence of the American people in the Government and reduce the Federal Government’s ability to address adequately vital public needs.” The first stated purpose of the act is to “improve the confidence of the American people in the capability of the Federal Government, by systematically holding Federal agencies accountable for achieving program results.”

The act mandates that all major decisions made by government agencies be properly justified in the public interest. One of the main outcomes of the act is that there is a much wider adoption of decision analysis and risk management in government organizations.

Private companies also understand the importance of decision analysis to justify their decisions. It is not enough for a company’s management to report to shareholders and Wall Street analysts that the company just spent X million dollars on research and development and Y million on capital projects. Investors need to see assurances that money was spent wisely. Therefore, many companies have started to establish structured decision analysis processes. Many organizations use decision support tools such as Enterprise Resource Management or Project Portfolio Management systems to improve their efficiencies. Six Sigma is a proven methodology to improve decision-making related to quality. One of the main areas of improvement, especially in the area of new product development, is the ability to successfully select which projects should go forward.

Government regulations and pressure from investors have become the driving forces behind the wider adoption of decision analysis. As government agencies and large companies implement the process, more information about decision analysis is becoming available, more experience is being accumulated, and more businesses are starting to improve the efficiencies of their projects though decision analysis.

A LITTLE BIT OF HISTORY

The fathers of decision analysis had lofty goals. In the 1700s, French-born mathematician Abraham de Moivre and English Presbyterian minister and mathematician Thomas Bayes tried to apply mathematics to prove the existence of God. Their work made important contributions to probabilities and statistics. In 1718 de Moivre published The Doctrine of Chances, in which he presented the concept of relative frequency for probabilities. De Moivre became one of fathers of the frequentistic approach to the theory of probabilities and statistics. Thomas Bayes came up with a different concept, one that would later become the foundation for the Bayesian theory in the field of probabilities. Around the same time, Swiss mathematician and physicist Daniel Bernoulli came up with idea of decision-making based on possible outcomes of events. Their work became the foundation of decision analysis.

The publication of the Theory of Games and Economic Behavior in 1944 by John von Neumann and Oskar Morgenstern was another significant step in decision science. After the theory was published, a number of scholars developed expansions and variations on it (Savage 1954; Luce 1959; Fishburn 1984; Karmarkar 1978; Payne 1973; Coombs 1975). Contemporary decision theory was introduced in the 1960s by Howard Raiffa and Robert Schlaifer of the Harvard Business School, who introduced the framework of decision analysis methods and tools (Raiffa 1968; Schlaifer 1969). The advent of the computer in the past few decades has also had a strong influence, and today decision and risk analysis software has become a practitioner’s tool.

Interestingly, in 2002 the Nobel Prize for economics was awarded to a psychologist rather than an economist. Daniel Kahneman was awarded the prize for “having integrated insights from psychological research into economic science, especially concerning human judgment and decision-making under uncertainty.” (The Sveriges Riksbank Prize 2002). The research, which Kahneman conducted with Amos Tversky and other psychologists, outlined the basic psychological foundation behind decision-making, significantly changing our understanding of human behavior. It affected not only our understanding of economics but also other areas, including project management.

DECISION ANALYSIS TODAY

Built on the work of many scholars in numerous fields, decision analysis has now become a practical framework that helps to solve many problems in different areas, including project management. The methodology is widely used by many companies—General Motors, DuPont, Boeing, Eli Lilly, AT&T, Exxon Mobil, Shell, Chevron, BP Amoco, Novartis, Baxter Bioscience, Bristol-Myers Squibb, and Johnson & Johnson, to name a few—and in government agencies such as the Department of Defense, Department of Homeland Security, and NASA. Because easy-to-use decision analysis software tools have become widely available, the adoption of decision analysis methods in all types of organizations, including small and medium-sized companies, has accelerated (see Appendix A).

Many universities, including Stanford University, Harvard University, Duke University, the London School of Economics and Political Science, UCLA, and the University of Massachusetts, offer courses in decision analysis. And a substantial number of scientific papers, textbooks, and reference works on the subject have been published in recent years (Keefer and Kirkwood 2004).

Experts in decision analysis have joined together in a number of professional organizations, one being the Decision Analysis Society. The Decision Analysis Society is a subdivision of the Institute for Operations Research and the Management Sciences (INFORMS). It publishes the journal Decision Analysis and has group meetings in conjunction with INFORMS annual meetings. Another professional group, the Decision Analysis Affinity Group (DAAG), focuses mostly on practical aspects of decision analysis. Project decision and risk analysis is a part of the agenda of the Risk Special Interest Group (RiskSIG) of the Project Management Institute (PMI). The Society for Judgment and Decision-making (SJDM) focuses mostly on the behavioral aspects of decision theory. DAAG, RiskSIG, and SJDM hold annual meetings where members and invited experts share their research and practical applications.