Efnisyfirlit

• Front Matter
• Contents
• Foreword
• Preface
• Accompanying website at http://www.wiley.com/college/goodwin/
• 1 Introduction
• Complex decisions
• The role of decision analysis
• Good and bad decisions and outcomes
• Applications of decision analysis
• Improved strategic decision making at Du Pont8
• Structuring decision problems in the International Chernobyl Project9, 10
• Selecting research projects at a large international pharmaceutical company11
• Petroleum exploration decisions at the Phillips Petroleum Company12
• Prioritizing infrastructure-renewal projects at MIT13
• Supporting the systems-acquisition process for the US military14
• Prioritizing projects in a busy UK social services department15
• Selecting a wide area network solution at EXEL Logistics16
• Planning under a range of futures in a financial services firm
• Supporting top-level political decision making in Finland17
• Automating advice-giving in a building society front office
• Allocating funds between competing aims in a shampoo-manufacturing company18
• Anticipating the need for doctors and dentists in the English National Health Service19
• Monitoring early warning signals in the business environment at Nokia and Statoil20
• The future of electric-drive vehicles in Germany21
• Overview of the book
• References
• 2 How people make decisions involving multiple objectives
• Introduction
• Heuristics used for decisions involving multiple objectives
• The recognition heuristic
• The minimalist strategy2
• Take the last2
• The lexicographic strategy8
• The semi-lexicographic strategy8
• Elimination by aspects12
• Sequential decision making: satisficing
• Reason-based choice
• Factors affecting which strategies people employ
• Other characteristics of decision making involving multiple objectives
• Decoy effects22
• Choosing by unique attributes
• Emotion and choice
• Justifying already-made choices
• Partitioning the total cost of an item changes preferences
• Summary
• Discussion questions
• References
• 3 Decisions involving multiple objectives: SMART
• Introduction
• Basic terminology
• Objectives and attributes
• Value and utility
• An office location problem
• An overview of the analysis
• Constructing a value tree
• Figure 3.1 – A value tree for the office location problem
• Measuring how well the options perform on each attribute
• Table 3.1 – Costs associated with the seven offices
• Direct rating
• Figure 3.2 – A value scale for office image
• Table 3.2 – Values and weights for the office location problem
• Value functions
• Figure 3.3 – Constructing a value function for office floor area
• Figure 3.4 – A value function for distance from customers
• Determining the weights of the attributes
• Figure 3.5 – Derivation of swing weights. For example, a swing from the worst to the best location for visibility is considered to be 80% as important as a swing from the worst to the best location for closeness to customers
• Aggregating the benefits using the additive model
• Trading benefits against costs
• Figure 3.6 – A plot of benefits against costs for the seven offices
• Sensitivity analysis
• Figure 3.7 – Sensitivity analysis for weight placed on image
• Theoretical considerations
• The axioms of the method
• Assumptions made when aggregating values
• Conflicts between intuitive and analytic results
• Value-focused thinking
• Summary
• Exercises
• References
• 4 Decisions involving multiple objectives: alternatives to SMART
• Introduction
• SMARTER
• Table 4.1 – ROC weights
• Even Swaps
• Even Swaps versus SMART
• The relative strengths of Even Swaps
• The relative limitations of Even Swaps
• The analytic hierarchy process
• Figure 4.1 – A hierarchy for the packaging machine problem
• Table 4.2 – Comparing the importance of ‘Costs’ and ‘Quality’
• Table 4.3 – Comparing the importance of the ‘Quality’ attributes
• Table 4.4 – Comparing the machines on ‘purchase cost’
• Figure 4.2 – Weights for the packaging machine problem
• Performing AHP calculations by hand
• Table 4.5 – Random indices for checking the consistency of a table
• The axioms of the AHP
• The AHP versus SMART
• The relative strengths of the AHP
• Criticisms of the AHP
• MACBETH
• Summary
• Exercises
• References
• 5 Introduction to probability
• Introduction
• Outcomes and events
• Approaches to probability
• The classical approach
• The relative frequency approach
• The subjective approach
• Mutually exclusive and exhaustive events
• The addition rule
• Table 5.1 – The frequency of flooding of a tidal river in April over the last 20 years
• Complementary events
• Marginal and conditional probabilities
• Table 5.2 – Results of a survey of workers in a branch of the chemicals industry
• Independent and dependent events
• The multiplication rule
• Probability trees
• Figure 5.1 – A probability tree
• Probability distributions
• Figure 5.2 – Probability distribution for the number of planes in service by the end of the year
• Figure 5.3 – Probability distribution for project completion time
• Figure 5.4 – Cumulative distribution function for project completion time
• Expected values
• The axioms of probability theory
• Summary
• Exercises
• 6 Decision making under uncertainty
• Introduction
• The maximin criterion
• Table 6.1 – A decision table for the food manufacturer
• The expected monetary value criterion
• Table 6.2 – Another decision table for the food manufacturer
• Figure 6.1 – A sensitivity analysis for the food manufacturer’s problem
• Limitations of the EMV criterion
• Table 6.3 – Returns and probabilities for the new component problem
• Single-attribute utility
• Figure 6.2 – A decision tree for the conference organizer’s problem
• Figure 6.3 – The conference organizer’s decision tree with utilities
• Figure 6.4 – A demonstration of how expected utility reduces the decision to a simple choice between lotteries
• Interpreting utility functions
• Figure 6.5 – A utility function for the conference organizer
• Figure 6.6 – Interpreting the shape of a utility function
• Utility functions for non-monetary attributes
• Figure 6.7 – A decision tree for the drug company research department problem
• Figure 6.8 – A utility function for product development time
• The axioms of utility
• Figure 6.9 – The continuity axiom
• Figure 6.10 – The substitution axiom
• Figure 6.11 – Demonstration of the substitution axiom
• Figure 6.12 – Demonstration of the unequal-probability axiom
• Figure 6.13 – Demonstration of the compound-lottery axiom
• More on utility elicitation
• How useful is utility in practice?
• Figure 6.14 – Allais’s paradox
• Multi-attribute utility
• The Decanal Engineering Corporation
• Figure 6.15 – A decision tree for the project manager’s problem
• Mutual utility independence
• Figure 6.16 – Determining utility independence
• Deriving the multi-attribute utility function
• Stage 1
• Figure 6.17 – Utility functions for overrun time and project cost
• Table 6.4 – The project manager’s utilities for overrun and cost
• Stage 2
• Figure 6.18 – Determining k1
• Figure 6.19 – Determining k2
• Figure 6.20 – The project manager’s decision tree with utilities
• Stage 3
• Figure 6.21 – Checking the consistency of the decision maker’s responses
• Interpreting multi-attribute utilities
• Further points on multi-attribute utility
• Summary
• Exercises
• References
• 7 Decision trees and influence diagrams
• Introduction
• Constructing a decision tree
• Figure 7.1 – An initial decision tree for the food-processor problem
• Figure 7.2 – A new decision tree for the food-processor problem
• Determining the optimal policy
• Figure 7.3 – Rolling back the decision tree
• Decision trees and utility
• Figure 7.4 – The engineer’s utility function
• Figure 7.5 – Apply the rollback method to a decision tree involving utilities
• Decision trees involving continuous probability distributions
• Figure 7.6 – The extended Pearson–Tukey (EP-T) approximation method
• Assessment of decision structure
• Figure 7.7 – One decision-analytic representation of the calculator problem
• Figure 7.8 – Toward the correct decision-analytic representation of the calculator problem?
• Figure 7.9 – Phases of a decision analysis
• Figure 7.10 – A possible fault tree for discovering why a car will not start (adapted from Fischhoff, B., Slovic, P. and Lichtenstein (1978) Fault Trees: Sensitivity of Estimated Failure Probabilities to Problem Representation, Journal of Experimental Psychology: Human Perception and Performance, 4(2), 330–344. Copyright 1978 © American Psychological Association. By permission of the authors)
• Eliciting decision-tree representations
• Figure 7.11 – Definitions used in influence diagrams
• Figure 7.12 – Influence diagram
• Figure 7.13 – Decision tree derived from influence diagram
• Summary
• Exercises
• References
• 8 Applying simulation to decision problems
• Introduction
• Monte Carlo simulation
• Table 8.1 – Ten simulations of monthly cash flows
• Table 8.2 – Estimating probabilities from the simulation results
• Table 8.3 – The effect of the number of simulations on the reliability of the probability estimates
• Applying simulation to a decision problem
• The Elite Pottery Company
• Stage 1: Identify the factors
• Figure 8.1 – Identifying the factors that will affect the profit earned by the commemorative plate
• Stage 2: Formulate a model
• Stage 3: Preliminary sensitivity analysis
• Table 8.4 – Estimates of lowest, highest and most likely values for the Elite Pottery problem
• Figure 8.2 – Tornado diagram showing the effect on profit if each factor changes from its lowest to its highest possible value
• Stage 4: Assess probability distributions
• Figure 8.3 – Probability distributions for variable costs, sales and fixed costs
• Stage 5: Perform the simulation
• Figure 8.4 – Probability distribution for profit earned by the commemorative plate
• Stage 6: Sensitivity analysis on the results of the simulation
• Stage 7: Compare alternative course of action
• Plotting the two distributions
• Figure 8.5 – A comparison of the profit probability distributions of the commemorative plate and the figurine
• Determining the option with the highest expected utility
• Stochastic dominance
• First-order stochastic dominance
• Figure 8.6 – First-order stochastic dominance
• Second-order stochastic dominance
• Figure 8.7 – Second-order stochastic dominance
• Figure 8.8 – An example where the test for second-order stochastic dominance is inconclusive
• The mean–standard deviation approach
• Figure 8.9 – The mean–standard deviation screening method
• Figure 8.10 – (a) A normal probability distribution for profit; (b) examples of quadratic utility functions
• Applying simulation to investment decisions
• The NPV method
• Table 8.5 – Calculating the NPVs for the Alpha and Beta machines
• Using simulation
• Figure 8.11 – Probability distributions for the Alpha machine (vertical axes represent probability density)
• Figure 8.12 – Probability distributions for the NPVs of the Alpha and Beta machines
• Utility and net present value
• Modeling dependence relationships
• Summary
• Exercises
• References
• 9 Revising judgments in the light of new information
• Introduction
• Bayes’ theorem
• Figure 9.1 – Tree diagram for the components problem
• Figure 9.2 – Applying Bayes’ theorem to the components problem
• Example
• Answer
• Figure 9.3 – Applying Bayes’ theorem to the equipment operating problem
• Another example
• Figure 9.4 – Applying Bayes’ theorem to the sales manager’s problem
• The effect of new information on the revision of probability judgments
• Figure 9.5 – The effect of vague prior probabilities and very reliable information
• Figure 9.6 – The effect of the reliability of new information on the modification of prior probabilities for the gas-exploration problem
• Applying Bayes’ theorem to a decision problem
• Figure 9.7 – (a) A decision tree for the retailer’s problem based on prior probabilities; (b) applying Bayes’ theorem to the retailer’s problem; (c) a decision tree for the retailer’s problem using posterior probabilities
• Assessing the value of new information
• The expected value of perfect information
• Figure 9.8 – Determining the expected value of perfect information
• Table 9.1 – Calculating the expected value of perfect information
• The expected value of imperfect information
• Figure 9.9 – Deciding whether or not to buy imperfect information
• Figure 9.10 – (a) Revising the prior probabilities when the test indicates that the virus is present; (b) revising the prior probabilities when the test indicates that the virus is absent
• Figure 9.11 – Determining the expected value of imperfect information
• Summary
• Exercises
• Reference
• 10 Heuristics and biases in probability assessment
• Introduction
• Test your judgment
• Heuristics and biases
• The availability heuristic
• Biases associated with the availability heuristic
• 1. When ease of recall is not associated with probability
• 2. Ease of imagination is not related to probability
• Test your judgment: answers to questions 1 and 2
• 3. Illusory correlation
• The representativeness heuristic
• Biases associated with the representativeness heuristic
• 1. Ignoring base-rate frequencies
• Test your judgment: answer to question 3
• 2. Expecting sequences of events to appear random
• Test your judgment: answer to question 4
• 3. Expecting chance to be self-correcting
• Test your judgment: answers to questions 5 and 6
• 4. Ignoring regression to the mean
• Test your judgment: answer to question 7
• 5. The conjunction fallacy
• Test your judgment: answers to questions 8 and 9
• The anchoring and adjustment heuristic
• Biases associated with anchoring and adjustment
• 1. Insufficient adjustment
• Test your judgment: answer to question 10
• 2. Overestimating the probability of conjunctive events
• Test your judgment: answer to question 11
• 3. Underestimating probabilities for disjunctive events
• Test your judgment: answer to question 12
• 3. Overconfidence
• Test your judgment: answer to question 13
• Other judgmental biases
• 1. Believing desirable outcomes are more probable
• 2. Biased assessment of covariation
• Test your judgment: answer to question 14
• Is human probability judgment really so poor?
• 1. Participants in studies may be unrepresentative of real decision makers
• 2. Laboratory tasks may be untypical of real-world problems
• 3. Experimental tasks may be understood in different ways by participants
• 4. Participants may be poorly motivated
• 5. Citation bias
• 6. Real-world studies suggest better performance
• 7. People think in terms of frequencies not probabilities
• Figure 10.1 – A methodology for choosing how to develop a subjective probability assessment
• Exercises
• References
• 11 Methods for eliciting probabilities
• Introduction
• Probability assessment
• Issues with verbal probability expressions
• Coherence in probability judgments
• Two barriers to improving probability assessments through learning
• Preparing for probability assessment
• Motivating
• Structuring
• Conditioning
• Assessment methods
• Assessment methods for individual probabilities
• Direct assessments
• The probability wheel
• Figure 11.1 – A probability wheel
• Assessment methods for probability distributions
• The probability method
• Graph drawing
• Figure 11.2 – The method of relative heights
• A comparison of the assessment methods
• Consistency and coherence checks
• Assessing the validity of probabilities
• Figure 11.3 – Calibration curves
• Assessing probabilities for very rare events
• Event trees
• Figure 11.4 – An event tree
• Fault trees
• Figure 11.5 – A fault tree
• Using a log-odds scale
• Figure 11.6 – A log-odds scale
• Communicating probability estimates
• Summary
• Exercises
• References
• 12 Risk and uncertainty management
• Introduction
• The Two Valleys Company
• Exploring sources of uncertainty
• Figure 12.1 – Sources of uncertainty at the Two Valleys Company
• Table 12.1 – Estimated values for uncertain factors
• Figure 12.2 – Cumulative probability distributions for annual profit at the two sites
• Identifying possible areas for uncertainty management
• 1. Calculate the effect of perfect control
• Table 12.2 – Results of risk management actions
• Figure 12.3 – Tornado diagram for Littleton
• 2. Repeat the above process for the next best option
• Figure 12.4 – Tornado diagram for Callum Falls
• Using brainstorming to create actions to improve the preferred policy
• Table 12.3 – Ideas for risk management at Two Valleys Company
• Figure 12.5 – The effectiveness of risk management measures
• Summary
• Exercise
• Figure 12.6 – Cumulative probability distribution for the AB Charity
• Figure 12.7 – Tornado diagram for the AB Charity
• References
• 13 Decisions involving groups of individuals
• Introduction
• Mathematical aggregation
• Table 13.1 – The production manager’s utilities and probabilities
• Table 13.2 – The accountant’s utilities and probabilities
• Table 13.3 – The average of the utilities and probabilities
• Aggregating judgments in general
• Taking a simple average of the individual judgments
• Taking a weighted average of the individual judgments
• Aggregating probability judgments
• Table 13.4 – Averaging probabilities
• Aggregating preference judgments
• Aggregating preference orderings
• Aggregating values and utilities
• Figure 13.1 – Measuring individuals’ strengths of preference against a common scale
• Unstructured group processes
• The Delphi method
• Figure 13.2 – The accuracy of the median estimate
• The benefits of establishing heterogeneity in Delphi panels
• Enhancing the evaluation of rationales by devil’s advocacy and dialectical inquiry
• Prediction markets
• Decision conferencing
• Summary
• Discussion questions
• References
• 14 Resource allocation and negotiation problems
• Introduction
• Modeling resource allocation problems
• An illustrative problem
• The main stages of the analysis
• Determining the areas, resources and benefits5
• Identifying the possible strategies for each region
• Figure 14.1 – Possible strategies identified by managers of the furniture company
• Assessing the costs and benefits of each strategy
• Table 14.1 – Values of the strategies in the individual regions
• Measuring each benefit on a common scale
• Figure 14.2 – The within-criterion weights for the furniture company problem
• Table 14.2 – Values of strategies with each benefit measured on a common scale
• Comparing the relative importance of the benefits
• Figure 14.3 – The across-criteria weights for the furniture company problem
• Identifying the costs and benefits of the packages
• Figure 14.4 – Identifying the efficient frontier for the furniture company problem
• Figure 14.5 – Investigating the costs and benefits of strategies in the individual regions
• Sensitivity analysis
• Negotiation models
• An illustrative problem
• Figure 14.6 – Management and union value functions and weights
• Table 14.3 – Calculation of values for the tentative management–union deal
• Figure 14.7 – Identifying the efficient frontier for the management–union negotiations
• Practical applications
• Summary
• Discussion questions
• Table 14.4 – Details of deals
• Figure 14.8 – Values of deals
• References
• 15 Decision framing and cognitive inertia
• Introduction
• Creativity in problem solving
• Figure 15.1 – Moves to attain the goal state in the first water-jug problem
• Figure 15.2 – Moves to attain the goal state in the second water-jug problem
• Figure 15.3 – The nine-dot problem
• How people frame decisions
• Solving the wrong problem
• Get hooked on complexity – overlooking simple options
• Imposing imaginary constraints and false assumptions on the range of options
• Sensitivity to reference points
• Prospect theory
• Figure 15.4 – Value function in prospect theory
• Figure 15.5 – Decision weights in prospect theory
• Figure 15.6 – Choice between alternatives P and Q
• Figure 15.7 – P and Q with values and decision weights replacing original numbers
• Mental accounting and the narrow bracketing of decisions
• Inertia in strategic decision making
• Real-world studies
• Studies in the psychological laboratory
• Non-rational escalation of commitment
• How people react to a threat
• Rethinking decisions
• Figure 15.8 – An example of a response to Russo and Schoemaker’s frame analysis questions
• Studies in the psychological laboratory and cognitive inertia: a synthesis
• Figure 15.9 – The relationship between the perceived business environment and the strategic decision process
• Summary
• Discussion questions
• Appendix
• Figure 15.10 – Solution to the nine-dot problem (Figure 15.3)
• References
• 16 Scenario planning: an alternative way of dealing with uncertainty
• Introduction
• Figure 16.1 – The Müller–Lyer illusion
• Scenario construction: the extreme-world method
• Figure 16.2 – Steps in scenario construction: the extreme-world method
• Figure 16.3 – Predetermined trends
• Figure 16.4 – Key uncertainties
• Figure 16.5 – Three scenarios
• Using scenarios in decision making
• Figure 16.6 – An illustrative business idea for a business school
• Figure 16.7 – Testing the robustness of strategies against scenarios
• Figure 16.8 – A ‘real’ scenario of future trading patterns
• Scenario construction: the driving forces method
• Figure 16.9 – An output of a ‘driving forces’ scenario structuring methodology
• Figure 16.10 – Adam Kahane’s four South African scenarios
• Figure 16.11 – Steps in scenario construction: the driving force method
• Figure 16.12 – Stakeholder structuring space
• Case study of a scenario intervention in the English National Health Service6
• Case study of a scenario intervention in the public sector7
• Forward to the past
• Free enterprise
• People’s kailyard8
• Technology serves
• Table 16.1 – The components of the futures methodologies
• Issues and limitations in scenario planning
• Organizational context and scenario planning
• Case study of an unsuccessful scenario-planning intervention
• Dealing with low predictability
• Conclusion
• Discussion questions
• References
• 17 Combining scenario planning with decision analysis
• Introduction
• Main stages of the approach
• Table 17.1 – Stages in the application of SMART to scenario planning
• Illustrative case study
• Stage 1: Formulate objectives
• Stage 2: Formulate scenarios
• Stage 3: Design alternative strategies
• Stage 4
• Table 17.2 – Strategy ranks across different scenarios for each objective (1 = best)
• Table 17.3 – Scores for strategies under the different scenarios for each objective
• Stage 5
• Table 17.4 – Ranking of swings
• Table 17.5 – Obtaining the weights for the objectives
• Stage 6: For each strategy/scenario combination use the performance scores and weights to determine a weighted aggregate score
• Table 17.6 – Obtaining an aggregate score for Status Quo in the Mail Mountain scenario
• Stage 7: Produce a table of strategy/scenario aggregate scores and use this to assess and compare the strategies’ performances, paying particular attention to the robustness of performance over the range of scenarios
• Table 17.7 – Aggregate scores
• Stage 8: Perform sensitivity analysis
• Table 17.8 – The effect of changing the weight on growth
• Extensions and alternatives to the above method
• Having a checklist of key objectives
• Making the ranking process easier
• Avoiding the need to assign scores and weights
• Different weights for different scenarios
• Other approaches
• Summary
• Exercises
• References
• 18 Alternative decision-support systems and conclusions
• Introduction
• Expert systems
• What is an expert system?
• What is expert knowledge?
• How is expert knowledge represented in expert systems?
• Marketing applications
• Financial services applications
• Figure 18.1 – Underwriting options
• Figure 18.2 – The rule base of the geographical module
• Where next?
• Back-office fraud-detection systems
• Point-of-sale advice-giving systems
• Conclusion
• Statistical models of judgment
• If linear models are so effective, why aren’t they more prevalent in practice?
• Comparisons of decision-aiding techniques
• Snap decisions and decision analysis: why not trust our initial intuitions?
• Designing decisions so that people make the ‘best’ choice
• Some final words of advice
• Are my assumptions about the business environment and the future valid?
• Who should I involve in the decision?
• Have I made an adequate search for alternative courses of action?
• How much effort is the decision worth? Which aspects of the problem require the most effort?
• Which method(s) are likely to help with my decision problem?
• Table 18.1 – Summary of techniques covered in the book
• Summary
• References
• Back Matter
• Suggested answers to selected questions
• Chapter 2
• Chapter 3
• Chapter 4
• Chapter 5
• Chapter 6
• Chapter 7
• Chapter 8
• Chapter 9
• Chapter 11
• Chapter 14
• Chapter 17
• Index