1. 1 Segment 4 Decision Making, Systems, Modeling, and Support

2. 2  Conceptual Foundations of Decision Making  The Systems Approach  How Support is Provided

3. 3 Typical Decision Aspects n Decision may be made by a group n Group member biases n Groupthink n Several, possibly contradictory objectives n Many alternatives n Results can occur in the future n Attitudes towards risk n Need information n Gathering information takes time and expense n Too much information n “What-if” scenarios n Trial-and-error experimentation with the real system may result in a loss n Experimentation with the real system - only once n Changes in the environment can occur continuously n Time pressure

4. 4 n How are decisions made??? n What methodologies can be applied? n What is the role of information systems in supporting decision making? DSS n Decision n Support n Systems,

5. 5 Decision Making n Decision Making: a process of choosing among alternative courses of action for the purpose of attaining a goal or goals n Managerial Decision Making is synonymous with the whole process of management

6. 6 System Input(s) Feedback Environment Output(s) Boundary Processes

7. 7 Environmental Elements Can Be n Social n Political n Legal n Physical n Economical n Often Other Systems

8. 8 An Information System n Collects, processes, stores, analyzes, and disseminates information for a specific purpose n Is often at the heart of many organizations n Accepts inputs and processes data to provide information to decision makers and helps decision makers communicate their results

9. 9 System Effectiveness and Efficiency Two Major Classes of Performance Measurement n Effectiveness is the degree to which goals are achieved Doing the right thing! n Efficiency is a measure of the use of inputs (or resources) to achieve outputs Doing the thing right! n MSS emphasize effectiveness Often: several non-quantifiable, conflicting goals

10. 10 Models n Major component of DSS n Use models instead of experimenting on the real system n A model is a simplified representation or abstraction of reality. n Reality is generally too complex to copy exactly n Much of the complexity is actually irrelevant in problem solving

11. 11 Degrees of Model Abstraction (Least to Most) n Iconic (Scale) Model: Physical replica of a system n Analog Model behaves like the real system but does not look like it (symbolic representation) n Mathematical (Quantitative) Models use mathematical relationships to represent complexity Used in most DSS analyses

12. 12 Benefits of Models 1. Time compression 2. Easy model manipulation 3. Low cost of construction 4. Low cost of execution (especially that of errors) 5. Can model risk and uncertainty 6. Can model large and extremely complex systems with possibly infinite solutions 7. Enhance and reinforce learning, and enhance training. Computer graphics advances: more iconic and analog models (visual simulation)

13. 13 The Modeling Process-- A Preview n How Much to Order for the Ma-Pa Grocery? n Bob and Jan: How much bread to stock each day? Solution Approaches n Trial-and-Error n Simulation n Optimization n Heuristics

14. 14 n Problem Decomposition: Divide a complex problem into (easier to solve) subproblems Chunking (Salami) n Some seemingly poorly structured problems may have some highly structured subproblems n Problem Ownership Outcome: Problem Statement

15. 15 The Design Phase n Generating, developing, and analyzing possible courses of action Includes n Understanding the problem n Testing solutions for feasibility n A model is constructed, tested, and validated Modeling n Conceptualization of the problem n Abstraction to quantitative and/or qualitative forms

16. 16 Mathematical Model n Identify variables n Establish equations describing their relationships n Simplifications through assumptions n Balance model simplification and the accurate representation of reality Modeling: an art and science

17. 17 Quantitative Modeling Topics n Model Components n Model Structure n Selection of a Principle of Choice (Criteria for Evaluation) n Developing (Generating) Alternatives n Predicting Outcomes n Measuring Outcomes n Scenarios

18. 18 Components of Quantitative Models n Decision Variables n Uncontrollable Variables (and/or Parameters) n Result (Outcome) Variables n Mathematical Relationships or n Symbolic or Qualitative Relationships

19. 19 n Decision n Uncontrollable Factors n Relationships among Variables Results of Decisions are Determined by the

20. 20 Result Variables n Reflect the level of effectiveness of the system n Dependent variables

21. 21 Decision Variables n Describe alternative courses of action n The decision maker controls them

22. 22 Uncontrollable Variables or Parameters n Factors that affect the result variables n Not under the control of the decision maker n Generally part of the environment n Some constrain the decision maker and are called constraints Intermediate Result Variables n Reflect intermediate outcomes

23. 23 The Structure of Quantitative Models n Mathematical expressions (e.g., equations or inequalities) connect the components n Simple financial model P = R - C n Present-value model P = F / (1+i) n

24. 24 Selection of a Principle of Choice Not the choice phase A decision regarding the acceptability of a solution approach n Normative n Descriptive

25. 25 Normative Models n The chosen alternative is demonstrably the best of all (normally a good idea) n Optimization process n Normative decision theory based on rational decision makers

26. 26 Rationality Assumptions n Humans are economic beings whose objective is to maximize the attainment of goals; that is, the decision maker is rational n In a given decision situation, all viable alternative courses of action and their consequences, or at least the probability and the values of the consequences, are known n Decision makers have an order or preference that enables them to rank the desirability of all consequences of the analysis

27. 27 Suboptimization n Narrow the boundaries of a system n Consider a part of a complete system n Leads to (possibly very good, but) non-optimal solutions n Viable method

28. 28 Descriptive Models n Describe things as they are, or as they are believed to be n Extremely useful in DSS for evaluating the consequences of decisions and scenarios n No guarantee a solution is optimal n Often a solution will be good enough n Simulation: Descriptive modeling technique

29. 29 Descriptive Models n Information flow n Scenario analysis n Financial planning n Complex inventory decisions n Markov analysis (predictions) n Environmental impact analysis n Simulation n Waiting line (queue) management

30. 30 Satisficing (Good Enough) n Most human decision makers will settle for a good enough solution n Tradeoff: time and cost of searching for an optimum versus the value of obtaining one n Good enough or satisficing solution may meet a certain goal level is attained

31. 31 Why Satisfice? Bounded Rationality n Humans have a limited capacity for rational thinking n Generally construct and analyze a simplified model n Behavior to the simplified model may be rational n But, the rational solution to the simplified model may NOT BE rational in the real-world situation n Rationality is bounded by –limitations on human processing capacities –individual differences n Bounded rationality: why many models are descriptive, not normative

32. 32 Developing (Generating) Alternatives n In Optimization Models: Automatically by the Model! Not Always So! n Issue: When to Stop?

33. 33 Predicting the Outcome of Each Alternative n Must predict the future outcome of each proposed alternative n Consider what the decision maker knows (or believes) about the forecasted results n Classify Each Situation as Under –Certainty –Risk –Uncertainty

34. 34 Decision Making Under Certainty n Assumes complete knowledge available (deterministic environment) n Example: U.S. Treasury bill investment n Typically for structured problems with short time horizons n Sometimes DSS approach is needed for certainty situations

35. 35 Decision Making Under Risk (Risk Analysis) n Probabilistic or stochastic decision situation n Must consider several possible outcomes for each alternative, each with a probability n Long-run probabilities of the occurrences of the given outcomes are assumed known or estimated n Assess the (calculated) degree of risk associated with each alternative

36. 36 Risk Analysis n Calculate the expected value of each alternative n Select the alternative with the best expected value n Example: poker game with some cards face up (7 card game - 2 down, 4 up, 1 down)

37. 37 Decision Making Under Uncertainty n Several outcomes possible for each course of action n BUT the decision maker does not know, or cannot estimate the probability of occurrence n More difficult - insufficient information n Assessing the decision maker's (and/or the organizational) attitude toward risk n Example: poker game with no cards face up (5 card stud or draw)

38. 38 Measuring Outcomes n Goal attainment n Maximize profit n Minimize cost n Customer satisfaction level (minimize number of complaints) n Maximize quality or satisfaction ratings (surveys)

39. 39 Scenarios Useful in n Simulation n What-if analysis

40. 40 Importance of Scenarios in MSS n Help identify potential opportunities and/or problem areas n Provide flexibility in planning n Identify leading edges of changes that management should monitor n Help validate major assumptions used in modeling n Help check the sensitivity of proposed solutions to changes in scenarios Decision

41. 41 Possible Scenarios n Worst possible (low demand, high cost) n Best possible (high demand, high revenue, low cost) n Most likely (median or average values) n Many more n The scenario sets the stage for the analysis

42. 42 The Choice Phase n The CRITICAL act - decision made here! n Search, evaluation, and recommending an appropriate solution to the model n Specific set of values for the decision variables in a selected alternative The problem is considered solved only after the recommended solution to the model is successfully implemented

43. 43 Search Approaches n Analytical Techniques n Algorithms (Optimization) n Blind and Heuristic Search Techniques

44. 44 Evaluation: Multiple Goals, Sensitivity Analysis, What-If, and Goal Seeking n Evaluation (with the search process) leads to a recommended solution n Multiple goals n Complex systems have multiple goals Some may conflict n Typically, quantitative models have a single goal n Can transform a multiple-goal problem into a single-goal problem

45. 45 Common Methods n Utility theory n Goal programming n Expression of goals as constraints, using linear programming n Point system n Computerized models can support multiple goal decision making

46. 46 Sensitivity Analysis n Change inputs or parameters, look at model results Sensitivity analysis checks relationships Types of Sensitivity Analyses n Automatic n Trial and error

47. 47 Trial and Error n Change input data and re-solve the problem n Better and better solutions can be discovered n How to do? Easy in spreadsheets (Excel) –What-if –Goal seeking

48. 48 What-If Analysis n Spreadsheet example of a what-if query for a staffing problem

49. 49 Goal Seeking n Backward solution approach n Example: What interest rate causes an the net present value of an investment to break even? n In a DSS the what-if and the goal-seeking options must be easy to perform

50. 50 The Implementation Phase There is nothing more difficult to carry out, nor more doubtful of success, nor more dangerous to handle, than to initiate a new order of things (Machiavelli, 1500s) *** The Introduction of a Change *** Important Issues n Resistance to change n Degree of top management support n Users’ roles and involvement in system development n Users’ training

51. 51 Alternative Decision Making Models n Paterson decision-making process n Kotter’s process model n Pound’s flow chart of managerial behavior n Kepner-Tregoe rational decision-making approach n Hammond, Kenney, and Raiffa smart choice method n Cougar’s creative problem solving concept and model n Pokras problem-solving methodology n Bazerman’s anatomy of a decision n Harrison’s interdisciplinary approaches n Beach’s naturalistic decision theories

52. 52 Naturalistic Decision Theories n Focus on how decisions are made, not how they should be made n Based on behavioral decision theory n Recognition models n Narrative-based models

53. 53 Recognition Models n Policy n Recognition-primed decision model

54. 54 Narrative-based Models (Descriptive) n Scenario model n Story model n Argument-driven action (ADA) model n Incremental models n Image theory

55. 55 Other Important Decision- Making Issues n Personality types n Gender n Human cognition n Decision styles

56. 56 Cognition n Cognition : Activities by which an individual resolves differences between an internalized view of the environment and what actually exists in that same environment n Ability to perceive and understand information n Cognitive models are attempts to explain or understand various human cognitive processes

57. 57 Cognitive Style n The subjective process through which individuals perceive, organize, and change information during the decision-making process n Often determines people's preference for human-machine interface n Impacts on preferences for qualitative versus quantitative analysis and preferences for decision-making aids n Affects the way a decision maker frames a problem

58. 58 Cognitive Style Research n Impacts on the design of management information systems n May be overemphasized n Analytic decision maker n Heuristic decision maker

59. 59 Decision Styles The manner in which decision makers n Think and react to problems n Perceive their –Cognitive response –Values and beliefs n Varies from individual to individual and from situation to situation n Decision making is a nonlinear process The manner in which managers make decisions (and the way they interact with other people) describes their decision style n There are dozens

60. 60 Some Decision Styles n Heuristic n Analytic n Autocratic n Democratic n Consultative (with individuals or groups) n Combinations and variations n For successful decision-making support, an MSS must fit the –Decision situation –Decision style

61. 61 n The system –should be flexible and adaptable to different users –have what-if and goal seeking –have graphics –have process flexibility n An MSS should help decision makers use and develop their own styles, skills, and knowledge n Different decision styles require different types of support n Major factor: individual or group decision maker

62. 62 The Decision Makers n Individuals n Groups

63. 63 Individuals n May still have conflicting objectives n Decisions may be fully automated

64. 64 Groups n Most major decisions made by groups n Conflicting objectives are common n Variable size n People from different departments n People from different organizations n The group decision-making process can be very complicated n Consider Group Support Systems (GSS) n Organizational DSS can help in enterprise-wide decision-making situations

65. 65 Summary n Managerial decision making is the whole process of management n Problem solving also refers to opportunity's evaluation n A system is a collection of objects such as people, resources, concepts, and procedures intended to perform an identifiable function or to serve a goal n DSS deals primarily with open systems n A model is a simplified representation or abstraction of reality n Models enable fast and inexpensive experimentation with systems

66. 66 n Modeling can employ optimization, heuristic, or simulation techniques n Decision making involves four major phases: intelligence, design, choice, and implementation n What-if and goal seeking are the two most common sensitivity analysis approaches n Computers can support all phases of decision making by automating many required tasks n Personality (temperament) influences decision making n Gender impacts on decision making are inconclusive n Human cognitive styles may influence human-machine interaction n Human decision styles need to be recognized in designing MSS