Decision Support Systems - Chapter 2: Decision-Making Systems, Models, and Support

Chapter 2Decision-Making Systems, Models, and SupportDecision Support Systems1© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangOutline1. Decision making2. Systems3. Models4. A preview of the modeling process5. Decision making: the Intelligence Phase6. Decision making: the Design Phase7. Decision making: the Choice Phase8. Evaluation: Multiple goals, sentivity analysis, what-if and goal seeking9.Decision making: the Implementation Phase10. How decision are supported.11. Human cognition and decision styles.2© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang1. Decision MakingDecision making is a process of choosing among alternative courses of actions for the purpose of attaining a goal or goals.The four phases of the decision process are:Intelligence DesignChoiceImplementationDecision making = problem solving ?3© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangTeam-Based Decision-MakingTeam-based decision makingIncreased information sharingDaily feedbackSelf-empowermentShifting responsibility towards teamsElimination of middle management4© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang2. SystemsStructure of a system:InputsProcessesOutputsFeedback from output to decision makerSeparated from environment by boundarySurrounded by environment InputProcessesOutputboundaryEnvironment5© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang6© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangSystem TypesClosed systemIndependentTakes no inputsDelivers no outputs to the environmentBlack BoxOpen systemAccepts inputsDelivers outputs to environment7© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangSystem effectiveness and efficiencyTwo major performance measures:Effectiveness is the degree to which goals are achieved. It is concerned with the outputs of a system.Efficientcy is a measure of the use of inputs (or resources) to achieve outputs.Effectiveness is doing the right thingEfficiency is doing the thing right8© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang3. Models Used for DSSA model is a simplified representation or abstraction of reality.Models are classified into 3 groups:IconicSmall physical replication of systemAnalogBehavioral representation of systemMay not look like systemQuantitative (mathematical)Demonstrates relationships between systems9© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang4. A preview of the modeling processThere are some ways to solve a problemTrial-and-error with the real systemSimulationOptimizationHeuristics10© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang Phases of Decision-MakingSimon’s original three phases:IntelligenceDesignChoiceHe added fourth phase later:ImplementationBook adds fifth stage:Monitoring11© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangPhases of decision-making process12© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang5. Decision-Making: Intelligence PhaseScan the environment Analyze organizational goalsCollect dataIdentify problemCategorize problemProgrammed and non-programmedDecomposed into smaller partsAssess ownership and responsibility for problem resolution13© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang6. Decision-Making: Design PhaseDevelop alternative courses of actionAnalyze potential solutionsCreate modelTest for feasibility Validate resultsSelect a principle of choiceEstablish objectivesIncorporate into modelsRisk assessment Criteria and constraints14© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangComponents of quantitative modelsDecision variables: describe alternative courses of actions.Result variables: indicates how well the system performs or attains its goals. Result variables are considered dependent variables.Uncontrollable variables or parameters. There are factors that affect the result variables but not under the control of the decision maker.Intermediate result variables15© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangAn example of modelingMBI Corporation makes special-purpose computers. A decision must be made: How many computers should be produced next month at the Boston plant? Two types of computers are considered: the CC-7 which requires 300 days of labor and $10000 in materials, and the CC-8, which requires 500 days of labor and $12000. The profit of each CC-7 is $8000, whereas that of each CC-8 is $12000.The plant has a capacity of 200000 working days per month, and the material budget is $8 million per month. Marketing requires that at least 100 units of CC-7 and at least 200 units of the CC-8 be produced each month. The problem is to maximize the company’s profits by determining how many units of CC-7 and how many units of CC-8 should be produced each month.16© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangModeling by Linear programmingDecision variables:X1 = units of CC-7 to be producedX2 = units of CC-8 to be producedResult variable:Total profit = Z. The objective is the maximize total profit: Z = 8000X1+12000X2Uncontrollable variables (constraints):Labor constraint: 300X1 + 500X2  200000Budget constraint: 10000X1 + 15000X2  8000000Marketing requirements: X1  100Marketing requirements: X2  200Optimal solution: X1 = 333.33, X2 = 200, Profit= $5066667.17© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangSelection of a principle of choice Principle of choiceA criterion that describes acceptability of a solution approach.There are two main principles of choice: normative and descriptiveNormative ModelsOptimizationEffect of each alternativeRationalizationMore of good things, less of bad thingsCourses of action are known quantityOptions ranked from best to worseSuboptimizationDecisions made in separate parts of organization without consideration of whole18© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangDescriptive ModelsDescribe how things are believed to beTypically, mathematically basedApplies single set of alternativesExamples:SimulationsWhat-if scenariosCognitive mapNarratives19© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangDeveloping (generating) AlternativesGeneration of alternativesMay be automatic or manualCan be a lengthy processTake time and cost moneyAlternatives can be generated with heuristicsOutcome measured by goal attainment20© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangGood enough or satisficingSatisficing is the willingness to settle for less than ideal.Form of suboptimization“Bounded rationality” (Simon’s idea)Limited human capacityLimited by individual differences and biasesBounded rationality is also why many models are descriptive rather than normative.21© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang7.Decision-Making: Choice PhaseDecision making with commitment to actDetermine courses of actionAnalytical techniquesAlgorithmsHeuristicsBlind searchesAnalyze for robustness 22© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang8. Evaluation: Multiple goals, sentivity analysis, what-if and goal seekingMultiple goals. Managers want to attain simultaneous goals, where some of them are conflicts.  goal programming.Sensitivity analysis. Sensitivity analysis attempts to assess the impact of a change in the input data or parameters on the proposed solution.What-if analysis. “What will happen to the solution if an input variable, an assumption, or a parameter value is changed?”EX: “What will be the market share if the advertisement budget increases by 5 percent?”Goal seeking. Goal seeking analysis calculates the values of inputs necessary to achieve a desired level of an output (goal). It is a backward solution approach.23© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang9. Decision-Making: Implementation PhasePutting solution to workVague boundaries which include:Dealing with resistance to changeUser trainingUpper management support24© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangSource: Based on Sprague, R.H., Jr., “A Framework for the Development of DSS.” MIS Quarterly, Dec. 1980, Fig. 5, p. 13.25© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang10. How Decisions are Supported Support for Intelligence PhaseAutomaticData MiningExpert systems, CRM, neural networksManualOLAPKMSReportingRoutine and ad hoc26© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangDecision Support SystemsSupport for Design PhaseFinancial and forecasting modelsGeneration of alternatives by expert systemRelationship identification through OLAP and data miningRecognition through KMSBusiness process models from CRM, RMS, ERP, and SCM27© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangDecision Support SystemsSupport for Choice PhaseIdentification of best alternativeIdentification of good enough alternativeWhat-if analysisGoal-seeking analysisMay use KMS, GSS, CRM, ERP, and SCM systems28© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangDecision Support SystemsSupport for Implementation PhaseImproved communicationsCollaborationTrainingSupported by KMS, expert systems, GSS29© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang11. Decision-Making In HumansTemperamentHippocrates’ personality typesMyers-Briggs’ Type IndicatorKiersey and Bates’ Types and MotivationsBirkman’s True ColoursGender30© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and LiangDecision-Making In HumansCognitive stylesWhat is perceived?How is it organized?SubjectiveDecision stylesHow do people think?How do they react?Heuristic, analytical, autocratic, democratic, consultative 31© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang32© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang