Decision analysis (DA) provides assistance in making logical, consistent
decisions under uncertainty. This book instructs readers in applying DA
to a wide range of project decisions. This is a wholly rewritten and
expanded successor to the best-selling prior editions. The entire asset
life cycle is covered, from an initial feasibility analysis, to the
project plan, to the post-project review, and on to a look-back analysis
of the capital investment decision. The book primarily uses a business
investment perspective, and assumes maximizing value for the project
owner is the objective. However, DA applies to all manner of decisions
for individuals, government, and non-profit organizations. DA is a
problem-solving process. There are three key features: 1) probabilities
and probability distributions express best judgments about risks and
uncertainties. 2) The organization has a decision policy expressed as a
single metric (the objective function). 3) Probabilities and outcome
values combine in the probability-weighting expected value calculation.
This book aims to make decision making clear, simple, and logical. A
clear decision policy can be elusive, and we offer suggestions for
making trade-offs among conflicting objectives. Converting the three
pillars of project management (cost, schedule and performance) into
project value equivalents makes the trade-offs clear. This book is
intended for serious PM students and practitioners. This is an essential
concepts and how-to book. The scope is quantitative analysis, from
project inception to post-project review. Project cost and schedule
modeling, in modest detail, is essential to feasibility analysis. A
general background in PM and corporate planning will be helpful. The
methods are quantitative and straightforward. We presume the reader is
comfortable with basic algebra. Experience with spreadsheets will be
helpful, as Microsoft® Excel® is the basis for several downloadable
calculation examples. The book has six pages of Suggested Reading
annotated references (plus footnote additions), over 500 Glossary
definitions, and over 2200 Index entries. Online supplements include
several whitepapers and other documents, example calculation
spreadsheets, color images of several important figures, four videos
(including a critical chain simulation), and the Utility Elicitation
Program (as a web app). Key topics include: Decision trees and Monte Carlo simulation for calculating outcome distributions and expected values • Probability concepts, including Bayes’ rule for value of information analysis • Popular probability distribution types and when they apply • Eliciting expert judgments, with attention to potential cognitive and motivational biases • Recognizing the three pillars
project in terms of project value • A 10-step decision analysis
process applicable to typical decisions • Project modeling concepts and
techniques, with special attention to risk drivers and other correlations • Deterministic and stochastic sensitivity analysis • Decision policy
that separates objectives, time value, and risk attitude • @RISK® with
Microsoft® Project for project schedule simulations under uncertainty •
Logical, consistent risk policy expressed as a utility function • Merge bias when task chains converge at a merge point • Tail estimate bias when estimating highly uncertain quantities • Optimizer’s curse, a portfolio forecasting bias • Winner’s curse, a bias characteristic of auctions • Overconfidence and other common cognitive biases • Using the best of critical chain and Monte Carlo simulation • Stochastic variance between a deterministic and a stochastic model • Modeling
risk and uncertainty using probabilities, probability distributions,
explicit formula relationships, correlation coefficients, risk drivers,
conditional branching, and rework cycles.