Instructor Kamran Ali Chatha Room No. 4 36, 4 th Floor, SDSB Building Office Hours TBA Email kamranali@lums.edu.pk Telephone 042 3560 8094 Secretary/TA TBA TA Office Hours TBA Course URL (if any) http://suraj.lums.edu.pk/~ro/ DISC 321 Decision Analysis Fall Semester 2016 COURSE BASICS Credit Hours 4 Lecture(s) Nbr of Lec(s) Per Week 2 Duration 110 minutes Recitation/Lab (per week) Nbr of Lec(s) Per Week Duration Tutorial (per week) Nbr of Lec(s) Per Week Duration COURSE DISTRIBUTION Core Elective Open for Student Category Close for Student Category Core MGS (Juniors & Seniors ), Open for MGS Juniors & Seniors in phase II COURSE DESCRIPTION (BRIEF) Decision Analysis is a branch of science that focuses on utilizing quantitative techniques for the purpose for making sound managerial decisions under various forms of constraints (economic, temporal and behavioral). This course exposes students to the concepts, methods and techniques of decision analysis to conceptualize real world managerial problems, analyze them and find workable solutions. The course covers topics such as: decision trees, decision making under uncertainty, value of information, risk analysis using Monte Carlo simulation, risk attitude, multi objective decisions, optimization models, matrix games, negotiation analysis to name a few. A real world project and written case analyses provide avenues for practical learning. COURSE DESCRIPTION (ELABORATE) Major objectives of this course are: (1) To understand basic concepts, methods and techniques of decision analysis; (2) To develop a capability to use quantitative techniques (in relation to decision analysis) for analyzing and solving real world managerial problems; (3) To have a hands on experience of developing spreadsheet models (using Microsoft Excel and an add on software namely Palisade Suite) for modeling and analyzing decisions; Decision Analysis / Science is a branch of science that focuses on utilizing quantitative techniques for the purpose for making sound managerial decisions under various forms of constraints (economic, temporal and behavioral) faced in the real world problems. These problems may belong to an organization s functional areas such as finance, operations, engineering, HRM and marketing functions etc. The problems may also be interdisciplinary in nature in which case function or discipline specific techniques when applied to solving these problems may not necessarily result into holistic or practical solutions. In such scenarios the techniques developed within the discipline of decision analysis may provide broader frameworks and concepts that render practical solutions to such problems. There are numerous examples in various disciplines where decision analysis concepts are needed for making sound decisions, for example in software engineering (e.g. decision about choosing one technology or process over the other), legal decisions (e.g., understanding the effects of economic pressures on attributions of responsibility), risk assessments (e.g., assessing risks of nuclear
power or missile tests), marketing (e.g. launching specific product in a market) and managerial decision making (e.g., correcting biases in the assessment of risk). The decision analysis concepts and frameworks are equally applicable in problems belonging to many other disciplines as well. Decision analysis relies heavily on decision theory which is concerned with identifying values of different alternatives, uncertainties involved, their utilities, and other issues relevant to a given decision, its rationality, and the resulting optimal decision. In order to exercise these concepts decision theory borrows some of the concepts from probability theory. In order to achieve aforementioned objectives two major steps have been taken while designing the course: (1) a number of real world case studies are used in order to better comprehend applicability of decision analysis concepts and techniques in real world problems. Extended class room discussions on case study analyses will be instrumental in understanding key issues pertaining to application, managerial concerns, and assumptions around the technique while focusing on the real world problem, (2) a number of lab sessions have been included in order to develop practical skills of configuring and using spreadsheets for decision analysis. COURSE PREREQUISITE(S) DISC 203 Probability and Statistics DISC 212 Introduction to Management Science (Students should possess basic knowledge of Probability / Statistics and calculus) COURSE LEARNING OBJECTIVES Major objectives of this course are: 1. To expose students basic concepts, methods and techniques of decision analysis; 2. To learn using quantitative techniques (in relation to decision analysis) for analyzing and solving real world 3. managerial problems; To have a hands on experience of developing spreadsheet models (using Microsoft Excel and an add on software namely Palisade Suite) for modeling and analyzing decisions UNDERGRADUATE PROGRAM LEARNING GOALS & OBJECTIVES General Learning Goals & Objectives Goal 1 Effective Written and Oral Communication Objective: Students will demonstrate effective writing and oral communication skills Goal 2 Ethical Understanding and Reasoning Objective: Students will demonstrate that they are able to identify and address ethical issues in an organizational context. Goal 3 Analytical Thinking and Problem Solving Skills Objective: Students will demonstrate that they are able to identify key problems and generate viable solutions. Goal 4 Application of Information Technology Objective: Students will demonstrate that they are able to use current technologies in business and management context. Goal 5 Teamwork in Diverse and Multicultural Environments Objective: Students will demonstrate that they are able to work effectively in diverse environments. Goal 6 Understanding Organizational Ecosystems Objective: Students will demonstrate that they have an understanding of Economic, Political, Regulatory, Legal, Technological, and Social environment of organizations. Major Specific Learning Goals & Objectives Goal 7 (a) Program Specific Knowledge and Understanding Objective: Students will demonstrate knowledge of key business disciplines and how they interact including application to real world situations. Goal 7 (b) Understanding the science behind the decision making process (for MGS Majors) Objective: Students will demonstrate ability to analyze a business problem, design and apply appropriate decision support tools, interpret results and make meaningful recommendations to support the decision maker
Indicate below how the course learning objectives specifically relate to any program learning goals and objectives. PROGRAM LEARNING GOALS AND OBJECTIVES Goal 1 Effective Written and Oral Communication Goal 2 Ethical Understanding and Reasoning Goal 3 Analytical Thinking and Problem Solving Skills Goal 4 Application of Information Technology Goal 5 Teamwork in Diverse and Multicultural Environments Goal 6 Understanding Organizational Ecosystems Goal 7 (a) Program Specific Knowledge and Understanding Goal 7 (b) Understanding the science behind the decision making process LEARNING OUTCOMES COURSE LEARNING OBJECTIVES To learn using quantitative techniques (in relation to decision analysis) for analyzing and solving real world managerial problems (Obj 2); To have a hands on experience of developing spreadsheet models (using Microsoft Excel and an add on software namely Palisade Suite) for modeling and analyzing decisions (Obj 3); To expose students basic concepts, methods and techniques of decision analysis (Obj 1);; To learn using quantitative techniques (in relation to decision analysis) for analyzing and solving real world managerial problems (Obj 2); To have a hands on experience of developing spreadsheet models (using Microsoft Excel and an add on software namely Palisade Suite) for modeling and analyzing decisions (Obj 3); To understand basic concepts, methods and techniques of decision analysis. COURSE ASSESSMENT ITEM Written Case Analyses. Group Project (Presentation). Written Case Analyses. Midterm Exam Final Exam Written Case Analyses. Group Project. Class Participation. Group Project. Quizzes Midterm Exam Final Exam Decision Analysis Process, and accompanying concepts, methods and techniques. Palisade Suite for conducting quantitative analyses. Capability to take managerial decisions. GRADING BREAKUP AND POLICY Written Cases Analyses / Assignment(s): 20% Home Work: Quiz(s): 10% (generally announced, occasionally unannounced)
Class Participation: 15% Attendance: Midterm Examination: 10% Project: 15% Final Examination: 30% The instructor has the right of re assigning 5% of the grading criteria. Class Participation Policy Class participation grading will be carried out as per the following rules: a) 0.5 for being absent from the class. b) 0.25 for attending the class. c) 0.5 to 0.7 for little participation in the class discussion (awarded for asking questions relevant to a discussion, describing case facts, giving an opinion or idea in relation to the discussion). d) 1.0 to 1.5 for good participation in the class discussion (awarded for giving a valid contradictory viewpoint or comprehensive argument or rationale behind a concept). e) 2.0 for very good participation in the class discussion (awarded for hitting multiple ds as mentioned above) f) 2.5 for excellent participation in the class discussion (awarded for bringing to the class and supporting with solid argument some concepts which even instructor does not know) Group Project Students will engage in a group project. The group size will be decided based on course enrollment. Students will identify a decision situation in an organization and apply course concepts thus formulating and analyzing the problem. Following this they will synthesize and suggest an appropriate solution to the problem. They will share their solution with the case study organization, and understand from company personnel the likely problems in implementing their solutions. The feedback obtained from the company personnel will be incorporated in the final project report. A detailed description on group project will be provided once the course starts. *** A few of the student projects will be shortlisted for converting into teaching case. Students will be asked if they are interested to convert their projects into teaching cases that will be published in an international conference / case journal and will make part of the DA course in the future. EXAMINATION DETAIL Midterm Exam Yes/No: Yes Combine Separate: Combine Duration: 3 Hours in the Lab Preferred Date: Exam Specifications: Closed Books / Open Notes Final Exam Yes/No: Yes Combine Separate: Combine Duration: 4 Hours in the Lab Exam Specifications: Closed Books / Open Notes
COURSE OVERVIEW S. NO SESSION TYPE 1. Class 2. 3. TOPIC CASES AND READINGS ASSIGNMENT QUESTIONS SESSION Introduction 4. Lab Using Spreadsheet for Probability & Probability Distributions 5. Class 6. Class 7. Class Influence Diagram and Payoff Table Decision Trees Readings: INTRODUCTION (1) PB Chapter 2: Modeling in a Problem Solving Framework (Sections 2.1, 2.2, 2.3, 2.4) (2) Learning by the Case Method, by Hammond, J.S. (HBS # 9 376 241) Readings: (1) AWZ Chapter 5: Probability and Probability Distribution (Sections 5.1, 5.2, 5.3, 5.4, 5.5, 5.6) (2) AWZ Chapter 6: Normal, Binomial, Poisson, and Exponential Distributions (Sections 6.1, 6.2, 6.3, 6.7) MODELING DECISIONS Case: Athens Glass Works Reading: CLEMEN Chapter Structuring Decisions pp43 65 Readings: (1) Decision Trees for Decision Making (2) CLEMEN Chapter 3: Structuring Decisions pp69 83 As two class sessions are devoted to this: Read PB Chapter 2 for the first class session. Read Learning by the Case Method for the second class session. Read specified sections before the lab session. Practice examples in the lab. Focusing just on the prices discussed by Christina Matthews and Robert Alexander, which price would you recommend, $2.15 or $2.36? What are various elements of a decision tree? How are decision trees analyzed? Decision Trees Case: Freemark Abbey 1. Assuming Mr. Jaeger chooses to harvest the Riesling grapes before the storm arrives, how much money will he make? 2. Assuming Mr. Jaeger chooses to leave the grapes on the vine, what is the probability that the grapes will end up with botrytis, and how much money will he make if that occurs? 3. Taking account of all the various possibilities, what should Mr. Jaeger do? OBJECTIVES Decision analysis and problemsolving. Revision of probability distributions. Developing Influence Diagrams and Payoff Tables. Developing and analyzing decision trees. Developing and analyzing decision trees.
8. Class 9. Class 10. Class Sensitivity Analysis Sensitivity Analysis Decision Making under Uncertainty Reading: CLEMEN Chapter 5: Sensitivity Analysis pp174 192 Case: Dhahran Roads (A) Reading: Cash Flow and Time Value of Money (SKIM) MODELING UNCERTAINTY Reading: CLEMEN Chapter 4: Making Choices pp111 145 1. What do you recommend regarding the proposed contract for the Dhahran Roads project? 2. Be sure that your recommendation acknowledges any key sources of risk in the conduct of the project and any negotiable parameters of the proposed contract. 3. Does sensitivity analysis change your decision when compared to the base case? The role of sensitivity analysis in decision modeling, analyzing and making. The role of sensitivity analysis in decision modeling, analyzing and making. Making decisions in probabilistic situations. 11. Lab Using Spreadsheet for Decision Trees 12. Class 13. Class 14. Class Decision Making under Uncertainty Value of Information Value of Information Reading: AWZ Chapter 7: Decision Making under Uncertainty, Section 7.2, 7.3. develop an understanding of various functions of PrecisionTree module that relate to making and analyzing decision trees using software. Solve problems 36 and 37 given at the end of the chapter. Case: George s T Shirts 1. What are the financial outcomes if Lassiter orders 5,000 T shirts? 7,500? 10,000? 2. How many T shirts should Lassiter order? MID TERM EXAM Reading: CLEMEN Chapter 12: Value of Information, pp496 509 Case: Integrated Siting Systems, Inc. 1. What do you recommend Ms. Scott of what decision should be taken? 2. How concerned should you be about the probability of the standard system not working? Making decision trees using a spreadsheet. Making decisions in probabilistic situations. The influence of the availability of information on the decision. The influence of the availability of information on the decision.
15. Lab Spreadsheet Modeling for Decision Making under Uncertainty 16. Lab Simulation 17. Modeling with Spreadsheets 18. Class 19. Class Monte Carlo Simulations Monte Carlo Simulations How far off would your assessment have to be before you would change your recommendation? Reading: AWZ Chapter 7: Decision Making under Uncertainty, Section 7.4, 7.5. Reading: AWZ Chapter 16: Introduction to Simulation Modeling, Sections 16.3, 16.4, 16.5, and 16.6. Readings: (1) CLEMEN Chapter 11: Monte Carlo Simulation pp 459 487. (2) Probability Distributions and Simulation (SKIM). Case: (1) Calambra Olive Oil (A) (2) Calambra Olive Oil (B) 3. What about reputation? Can you afford the chance of such a visible failure? How much does reputation have to be worth to change the decision on economic grounds? 4. What is this test worth to you? What would you pay for a perfect information? Practice examples 7.2, 7.3, 7.4 in the lab. Practicing these examples will help you solve the following assignment. ASSIGNMENT: Solve problems 19, 21, 22 individually and submit your solutions. Read specified material before the lab. Practice examples 16.1, 16.2, 16.3, 16.4 and 16.5 in the lab. Practicing these examples will help you solve the following assignment. ASSIGNMENT: Solve problems 11, 17, 22, 26 individually and submit your solutions. The goal of this assignment is for you to help Frank Lockfeld better understand the risks he is taking with his Calambra Olive Oil venture and to help him figure out how many gallons of olive oil he should order in 1994. We will analyze this business problem in two parts. Part A: In the first part, you should use the spreadsheet model LIQUIDGOLD.XLS [available] and the ranges provided by Frank Lockfeld to develop a tornado Practicing probabilistic decisions using spreadsheets. Understanding RISK as a package to model decisions using simulations. Understanding Monte Carlo simulation method. Applying Monte Carlo Simulation method in a reallife business problem.
20. Lab Simulation 21. Modeling with Spreadsheets 22. Class Risk Attitude 23. Lab Incorporating Risk Attitude 24. Class Risk Attitude chart to identify the important uncertainties in the problem. Be sure you can explain any surprising findings in this analysis. Reading: AWZ Chapter 17: Simulation Models, Sections 17.2, 17.3, 17.4. MODELING PREFERENCES Reading: CLEMEN Chapter 13: Risk Attitude, pp 527 555. Reading: AWZ Chapter 7: Decision Making under Uncertainty, Section 7.6. Case: Risk Analysis for Merck & Company: Product KL 798 Part B: After that class you will be given the (B) case, which contains additional information about the uncertainties you identified as important. Using this information, you should develop a simulation model to resolve the key questions of the case: How much olive oil should Frank Lockfeld order? How risky is this venture? Practice examples 17.1, 17.2, 17.3, 17.4, 17.5, 17.7, 17.8, 17.9 in the lab. Practicing these examples will help you solve the following assignment. ASSIGNMENT: Solve problems 17, 18, 20 individually and submit your solutions. Practice example 7.5 in the lab. Solve problems 77, 79, 80 in the lab. For questions 1 and 2 only, assume that Merck will follow the advice of George W. Merck, We try never to forget that medicine is for the people. It is not for the profits. The profits follow, and if we have remembered that, they have never failed to appear. 1. First, do a risk neutral analysis. (a) Draw an influence diagram that captures the dynamics of the KL 798 opportunity. (b) What is the expected monetary value of the KL 798 opportunity? Be very clear about how your spreadsheet works. 2. Now consider risk aversion in your analysis. What would be the Understanding RISK as a package to model decisions using simulations. Understanding the influence of manager risk attitude on decisions. Practicing risk attitude using a spreadsheet. Understanding the influence of manager risk attitude on decisions.
25. Class 26. Class Structuring Multi Objective Decisions Additive Utility Function certainty equivalent for the total opportunity? Reading: CLEMEN Chapter 15: Conflicting Objectives I: Fundamental Objectives and the Additive Utility Function pp599 621. Case: Sleepmore Mattress Manufacturing; Plant Consolidation. 3. Assume that we ignore George W. Merck s advice and always seek the financially best path. (a) Draw a decision tree of the sequence of decisions and uncertainties and integrate it with the influence diagram from question 1a. (b) What now is the expected monetary value of KL 798? Clearly describe how you arrived at this solution and provide information on how your spreadsheet model works. (c) What is the certainty equivalent of KL 798 to Merck when considering their risk preference? 1. Be prepared to discuss all the decision approaches described in the note and consider how the approaches might be applied to the case. 2. Rate the four quantitative attributes, determine the appropriate weights for the attributes and compare the three locations. If you had to phase in the consolidations one at a time, in what order would you do them? 3. How sensitive is your ranking to the weights you assigned? 4. How would you score plant size at site 1 if the sales were $30 million at plant A rather than $3 million? Would you change the range of the scale, or the weight of the attribute, or both? 5. Implicit in your analysis are some trade offs that can be calculated. For example, what is the dollar value (in terms of initial cost) of improving the labor attribute by 1 unit on the 10 point scale? Understanding multi objective decisions and structuring them. Additive utility function as a method of analyzing multiobjective decisions.
27. Class 28. Class 29. Multi attribute Utility Models Reading: CLEMEN Chapter 16: Conflicting Objectives II: Multi attribute Utility Models with Interactions pp644 659 Project presentations (mandatory attendance by all students) Multiplicative utility function as a method of analyzing multiobjective decisions. FINAL EXAM TEXTBOOK(S)/SUPPLEMENTARY READINGS Following books are recommended for this course however, students are strongly encouraged to consult any other resources such as: books, journals, magazines, sharing personal experiences to enhance their learning. [AWZ]: Albright, S.C., Winston, W.L., and Zappe, C., 2006, Data Analysis & Decision Making With Microsoft Excel, 3e, Thomson, South Western, ISBN: 0 324 40083 7. [CLEMEN]: Clemen, R. T., 2001, Making Hard Decisions: An Introduction to Decision Analysis with Decision Tools, Duxbury Press, Thomson Learning, ISBN: 0 534 36597 3. [PB]: Powell, S.G., and Baker, K.R., 2009, Management Science The Art of Modeling with Spreadsheets, John Wiley & Sons Inc., ISBN 13: 978 0 470 39376 5. [ASW] Anderson, Sweeney & Williams, Statistics for Business and Economics.