Bài giảng Operations Management - Supplement 5 Decision Theory

Decision TheorySupplement 5Copyright © 2015 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.You should be able to:LO 5s.1 Outline the steps in the decision processLO 5s.2 Name some causes of poor decisions LO 5s.3 Describe and use techniques that apply to decision making under uncertaintyLO 5s.4 Describe and use the expected-value approachLO 5s.5 Construct a decision tree and use it to analyze a problemLO 5s.6 Compute the expected value of perfect informationLO 5s.7 Conduct sensitivity analysis on a simple decision problemSupplement 5: Learning ObjectivesSteps:Identify the problemSpecify objectives and criteria for a solutionDevelop suitable alternativesAnalyze and compare alternativesSelect the best alternativeImplement the solutionMonitor to see that the desired result is achievedDecision ProcessLO 5s.1Decisions occasionally turn out poorly due to unforeseeable circumstances; however, this is not the norm.More frequently poor decisions are the result of a combination ofMistakes in the decision processBounded rationalitySuboptimizationCauses of Poor DecisionsLO 5s.2Errors in the Decision ProcessFailure to recognize the importance of each stepSkipping a stepFailure to complete a step before jumping to the next stepFailure to admit mistakesInability to make a decisionMistakes in the Decision ProcessLO 5s.2There are three general environment categories:CertaintyEnvironment in which relevant parameters have known valuesRiskEnvironment in which certain future events have probabilistic outcomesUncertaintyEnvironment in which it is impossible to assess the likelihood of various possible future eventsDecision EnvironmentsLO 5s.3Decisions are sometimes made under complete uncertainty: No information is available on how likely the various states of nature are.Decision Criteria:MaximinChoose the alternative with the best of the worst possible payoffsMaximaxChoose the alternative with the best possible payoffLaplaceChoose the alternative with the best average payoff Minimax regretChoose the alternative that has the least of the worst regretsDecision Making Under UncertaintyLO 5s.3Decisions made under the condition that the probability of occurrence for each state of nature can be estimatedA widely applied criterion is expected monetary value (EMV)EMVDetermine the expected payoff of each alternative, and choose the alternative that has the best expected payoffThis approach is most appropriate when the decision maker is neither risk averse nor risk seekingDecision Making Under RiskLO 5s.4Possible Future DemandAlternativesLow (.30)Moderate (.50)High (.20)Small Facility$10$10$10Medium Facility71212Large Facility(4)216Example – EMV EMVsmall = .30(10) +.50(10) +.20(10) = 10EMVmedium = .30(7) + .50(12) + .20(12) = 10.5EMVlarge = .30(-4) + .50(2) + .20(16) = $3Build a medium facilityLO 5s.4Composed ofNodesDecisions – represented by square nodesChance events – represented by circular nodesBranchesAlternatives – branches leaving a square nodeChance events – branches leaving a circular nodeAnalyze from right to leftFor each decision, choose the alternative that will yield the greatest returnIf chance events follow a decision, choose the alternative that has the highest expected monetary value (or lowest expected cost)Decision TreeLO 5s.5A manager must decide on the size of a video arcade to construct. The manager has narrowed the choices to two: large or small. Information has been collected on payoffs, and a decision tree has been constructed. Analyze the decision tree and determine which initial alternative (build small or build large) should be chosen in order to maximize expected monetary value.Example – Decision Tree122$40$40$50$55($10)$50$70Build SmallLow Demand (.40)Low Demand (.40)High Demand (.60)High Demand (.60)Build LargeDo NothingCut PricesDo NothingOvertimeExpandLO 5s.5Example – Decision Tree122$40$40$50$55($10)$50$70Build SmallLow Demand (.40)Low Demand (.40)High Demand (.60)High Demand (.60)Build LargeDo NothingCut PricesDo NothingOvertimeExpandEVSmall = .40(40) + .60(55) = $49EVLarge = .40(50) + .60(70) = $62Build the large facilityLO 5s.5Expected value of perfect information (EVPI)The difference between the expected payoff with perfect information and the expected payoff under riskTwo methods for calculating EVPIEVPI = expected payoff under certainty – expected payoff under riskEVPI = minimum expected regretExpected Value of Perfect InformationLO 5s.6Sensitivity analysisDetermining the range of probability for which an alternative has the best expected payoffThe approach illustrated is useful when there are two states of natureIt involves constructing a graph and then using algebra to determine a range of probabilities over which a given solution is best.Sensitivity AnalysisLO 5s.7Sensitivity AnalysisLO 5s.7State of NatureAlternative#1#2SlopeEquationA41212 – 4 = +84 + 8P(2)B1622 – 16 = -1416 – 14P(2)C1288 - 12 = -412 – 4P(2)