DECISION MAKING, SYSTEMS, MODELING, AND SUPPORT Chapter 2 DECISION MAKING, SYSTEMS, MODELING, AND SUPPORT
Phases of the decision process Intelligence Design Choice Implementation Simon Huber
What is the relevance of this model to DSS? What are some of the challenges in each phase /sub-steps shown in Fig 2.1? Herbert Simon
Issues in Intelligence Phase Example: Data collection Data is not available or too much data Obtaining data is expensive Data may not be accurate /precise enough Data is qualitative; representing it is difficult Similarly, even though Simon’s model describes the general steps humans go through in decision-making, accomplishing each step can itself be complicated.
Classification & Decomposition (Intelligence phase) Problem classification - places problem in a definable category Egs: Product-mix, Capital budgeting, Negotiation - leads to a standard solution (canned) approach Problem decomposition - divide and conquer Eg: CSU: Common Management System (SOLAR)
Modeling for Decision Support (Design phase) Modeling involves conceptualizing a real-world problem and abstracting it to a quantitative form or qualitative form Models capture selected decision variables and their relationships
Types of Models Iconic (Scale): physical replica, Eg. Airplane, building Analog: symbolic representation of reality, Eg. Car dashboard, hierarchy chart Quantitative / Mathematical: uses analytic approach, Eg. LP, EOQ, Regression Descriptive / Mental: narrative, uses heuristics (jury deliberations), cognitive map (banxia.com), simulation/scenarios
Decision Making: (The Design Phase ) Measuring outcomes The value of an alternative is evaluated in terms of goal attainment Validate the model Done typically through historical data or pilot testing of the model over a short window of time
Bounded Rationality (Design phase) All alternatives will be evaluated Will look for the best (optimum) solution Bounded rationality: Sub-optimization: failure to look for an overall solution for the organization Satisficing (or good enough) solution Humans like to simplify problems consider fewer alternatives, criteria, constraints they are under time pressure, cost they have limited processing power
Decision Making: (The Choice Phase) Objective is to select an alternative/ reach a decision Perform Sensitivity analysis - A study of the effect of a change in one or more input variables on the proposed solution What-if analysis A process that involves asking a computer what the effect of changing some of the input data or parameters would be Scenario and Risk analysis Assess level of risk to the outcome associated with each potential alternative being considered
Decision Making: The Implementation Phase Generic implementation issues include: Resistance to change Degree of support of top management User training
How does DSS support Simon’s model of decision-making? Support for the intelligence phase The ability to scan external and internal information sources for opportunities and problems and to interpret what the scanning discovers Web tools and sources are extremely useful for environmental scanning Internal data sources/warehouses be scanned via a corporate intranet Set up agents/ triggers in software (eg. OS, SQL Server)
How does DSS support Simon’s model of decision-making? Support for the design phase Mostly human intelligence/effort OLAP, data mining to discover data relationships Cognitive mapping software Computational tools/ management science models Any existing ES/KMS in the decision topic
How does DSS support Simon’s model of decision-making? Support for the choice phase DSS can support through comparison of measurable outcomes of various alternatives; eg. Risk indexes, what-if (scenarios) and goal-seeking analyses (Excel spread-sheeting) KMS help explain heuristics / logic of decision steps A GDSS can provide support for group think that lead to consensus DSS support to Design & Choice phases overlap (See Fig 2.2)
How does DSS support Simon’s model of decision-making? Support for the implementation phase DSS can be used in implementation activities such as identifying tasks to be completed, critical path analysis, decision communication among team members, project management DSS can also help with training in the new system (many DSS software like SPSS come with tutorial modules)