Risk and Decision Analysis in Projects
Some of Schuyler's tried-and-true tips include: - The single-point estimate is almost always wrong, so that it is always better to express judgments as ranges. A probability distribution completely expresses someone's judgment about the likelihood of values within the range.- We often need a single-value cost or other assessment, and the expected value (mean) of the distribution is the only unbiased predictor. Expected value is the probability-weighted average, and this statistical idea is the cornerstone of decision analysis.- Some decisions are easy, perhaps aided by quick decision tree calculations on the back of an envelope. Decision dilemmas typically involve risky outcomes, many factors, and the best alternatives having comparable value. We only need analysis sufficient to confidently identify the best alternative. As soon as you know what to do, stop the analysis!- Be alert to ways to beneficially change project risks. We can often eliminate, avoid, transfer, or mitigate threats in some way. Get to know the people who make their living helping managers sidestep risk. They include insurance agents, partners, turnkey contractors, accountants, trainers, and safety personnel.
What people are saying - Write a review
We haven't found any reviews in the usual places.
LIST OF FIGURES
percent during 19952000 As this edition was being written
40 other sections not shown
Activities Complete alternative Appendix assessments base beta distributions calculations cash flow chance events Chapter component contingency continuous distribution correlation coefficient Crane Critical Chain critical path cumulative curve deci decision analysis decision maker decision model decision policy decision problems decision rule decision tree analysis delay detail discrete distribution EMV decision estimates evaluation example expected monetary value expected value forecast formula frequency histogram graph impact important influence diagram input variables investment judgments logical mean method million monetary value Monte Carlo simulation nodes normal distribution objective optimization outcome value parameter percent performance PMBOK possible outcomes probability distributions problem Project Cost project management project model project plan random represent risk event risk policy sampling scenario schedule sensitivity analysis shows single value solve spreadsheet statistic stochastic model stochastic variance technique tion trial typically usually utility function value measure Wastewater Plant