SBM-NMIMS: COURSE TEACHING PLAN Assurance of Learning AOL Specific Course Code Course Title Modelling and Optimization for Business Decisions Mumbai Campus:Dr. Abhinav Sharma (Course Anchor), Dr. T. Kachwala, Dr. Ashu Sharma, Dr. Akshay Khanzode, Prof. Arti Deo Bengaluru Campus: Prof. Satish Kumar S Course Hyderabad Campus: Dr Vengala Rao Pachava Instructor/s Indore Campus: Dr. Shubhangi Jore Navi Mumbai Campus: Prof. Prashant Barsing Credit 3 Value Programme FT MBA; Trim II & Trimester Pre-requisite CLO 1: Develop an understanding of various mathematical formulation techniques and models commonly used to resolve business problems. (PLO 3b) Learning Objectives Learning Outcomes (Must be connected to Learning Objectives) Course Description Evaluation Pattern CLO 2: Apply analytical techniques to solve the mathematical model of a business problem and derive solution(s). (PLO 4d) CLO 3: Analyze a business problem based on given information and data, and decide best course of action. (PLO 2c) By the end of this course: 1. Students will be able to develop a process view of business situations and come up with mathematical models. (CLO 1) 2. Students will be able to apply analytical skills to derive the best possible solutions to different types of quantitative problems. (CLO 2) 3. Students will be able to evaluate alternatives and choose the best one based on various evaluation criteria. (CLO 3) This course provides students with essential skills to address the complex decision-making challenges that managers encounter in the business world. The course focuses on a structured approach to identify, model, and solve different types of business problems, leading to effective resource utilization and competitive advantage. Students will learn how to apply various analytical tools and techniques, including forecasting, linear programming, and decision analysis, to enhance their understanding of business problems and generate optimal solutions. By the end of this course, students will have a deeper understanding of the problem-solving process and will be well-equipped to make data-driven decisions in realworld business situations. Specific assessment 3 Credit 1.5 Credit AOL Instruments CLO 1 CLO 2 CLO 3 1|Page methods Test 25 Project Work Class participation Final Exam 25 Total 100 (*) Embedded questions Rubric 25 10 10 5 15 10 15 25 45 20 10 Embedded questions 40 50 2 *AOL Assessment Instruments: Embedded Questions: Quiz, Class Test, Midterm Examination, Final Examination Rubrics: Case & Article Discussion, Individual Assignment Group Projects & Viva’s, Case Problem analysis, Oral and written communication presentations, Role Play, Group Presentation, Group Project etc. Pedagogy adopted for class engagement Learning Outcomes session wise Introduction to Pre-read: Chapter 15, pp. Class discussion, lecture, and modelling and 654 – 671 [Textbook] numerical problems. optimization for business problems. Learning outcome: Topics / Sub -topics Sessions 1 Forecasting: Introduction to time series analysis, importance of forecasting, common time series patterns, model evaluation metrics (MFE, MAE, MSE, MAPE), smoothing methods – moving average, weighted moving average Chapter detail / Article Reference / Case Studies 1. Students will be able to develop forecasting models with time series smoothing methods. 2. Students will demonstrate their ability to apply smoothing methods to make accurate forecasts for time series data with random variations only. 3. Students will be able to critically evaluate and compare the effectiveness of different forecasting models using a variety of metrics, such as Mean Forecast Error (MFE), Mean Absolute Error (MAE), Mean Squared Error (MSE), and Mean Absolute Percentage Error (MAPE) Forecasting: Exponential Pre-read: Chapter 15, pp. Class discussion, lecture, and smoothing method. 672 – 684 [Textbook] numerical problems. 2 Forecasting in presence Case: Roychowdhury, S., Learning outcome: of trend and seasonality: Shrivastava, A., and Dinesh Forecasting with 1. Students will demonstrate 2|Page seasonality without trend, Kumar, U. (2014). Forecasting with Forecasting Demand for seasonality and trend using Food at Apollo Hospitals. additive model Indian Institute of Management Bangalore. Forecasting in presence of trend and seasonality: Forecasting in presence of trend using Holt’s method, Time series decomposition for multiplicative model 3 their capacity to construct and analyze forecasting models for time series data exhibiting trends and/or seasonality, using Ordinary Least Squares (OLS) regression. Pre-read: Chapter 13, pp. Class discussion, lecture, and 378 [Reference book-1] numerical problems. Chapter 6, pp. 306 – 311 Case discussion questions: [Reference book-2] 1. Students will be able to use Case: Roychowdhury, S., Holt's method of forecasting to Shrivastava, A., and Dinesh generate accurate predictions Kumar, U. (2014). for relevant scenarios. Forecasting Demand for Food at Apollo Hospitals. 2. Students will exhibit the capability to methodically deIndian Institute of compose given time series data, Management Bangalore. isolating and identifying its random, trend, and seasonal components. Auto-regressive moving average (ARMA) and auto-regressive integrated moving average (ARIMA) processes: stationary vs non-stationary time series, AR process, MA process, ARMA process 4 Pre-read: Chapter 13, pp.388 – 398 [Reference book-1] Case: Roychowdhury, S., Shrivastava, A., and Dinesh Kumar, U. (2014). Forecasting Demand for Food at Apollo Hospitals. Indian Institute of Management Bangalore. Class discussion, lecture, and numerical problems. Class discussion: 1. What is need for ARMA and ARIMA models? Learning outcome: 1. Students will comprehend and understand the differences between stationary and nonstationary time series. 2. Students will demonstrate the capacity to construct and analyze (ARMA) models for specified time series data. 5 Pre-read: Chapter 13, pp.398 – 405 [Reference book-1] Case: Roychowdhury, S., Shrivastava, A., and Dinesh Kumar, U. (2014). Forecasting Demand for Power of forecasting Food at Apollo Hospitals. Institute of model: Theil’s coefficient Indian Auto-regressive moving average (ARMA) and auto-regressive integrated moving average (ARIMA) processes: ARIMA model building Class discussion, lecture, and numerical problems. Learning outcome: 1. Students will be able to determine the stationarity of the series using Augmented Dickey Fuller test (whichever is applicable). 3|Page Management Bangalore. 2. Students will demonstrate the ability to construct and analyze (ARIMA) models for a given time series, employing their understanding of ARIMA properties to effectively interpret the results. 3. Students will be able to estimate power of forecasting model using Theil’s coefficient. Pre-read: Chapter 2, pp. 28 Class discussion, lecture, and Linear programming: – 32 [Textbook] numerical problems Introduction to linear programming, formulation Case: Dhebar A. (1989). Merton Truck Company. Class discussion: of a linear program Harvard Business 1. The 4 stages in formulation Publishing Education and the need for these 4 stages. 2. The assumption of linearity and how this is useful. Is it wrong to make these assumptions in practice? 6 3. Assumption of continuity and how this is not a limitation. Learning Outcome: 1. Students will be capable of formulating a linear program for a given problem, utilizing and synthesizing the provided information to establish an effective mathematical model. Linear programming: Graphical solution, feasible area, optimal solution, infeasibility, unboundedness, redundancy, multiple optima. 7 Pre-read: Chapter 2, pp. 33 Class discussion, lecture, and – 46 [Textbook] numerical problems Case discussion questions: Case: Dhebar A. (1989). Q.1 Develop a linear program Merton Truck Company. to determine optimal productHarvard Business mix for Merton. Publishing Education Q.2 Determine using graphical method what should be the optimal product-mix for Merton. Learning Outcome: 1. Students will demonstrate the ability to derive the optimal solution to a linear program, employing the graphical 4|Page procedure to visually analyze and solve the problem. 2. Students will gain knowledge and understanding of the key concepts in linear programming such as infeasibility, unboundedness, redundancy, and multiple optima, and will be able to interpret the practical implications of these concepts when encountered in different scenarios. Linear programming: Pre-read: Chapter 17 Class discussion, lecture, and Limitations of graphical [Textbook, available online numerical problems method, Simplex method on book website] Case discussion questions: to solve linear programs Q.1 Using simplex method Case: Dhebar A. (1989). determine what will be the Merton Truck Company. optimal product-mix for Harvard Business Merton? Publishing Education 8 Learning Outcome: Linear programming: Sensitivity analysis with single and multiple coefficient changes 9 1. Students will demonstrate the ability to derive the optimal solution to a linear program by proficiently applying the Simplex method, utilizing their understanding of this key technique in linear programming. Pre-read: Chapter 3, pp. 94 Class discussion, lecture, and – 110 [Textbook] numerical problems Chapter 15, pp. 456 – 457 Case discussion questions: [Reference book 1] Q.1 What would be the best product mix if enginer Case: Dhebar A. (1989). assembly capacity were raised Merton Truck Company. by one unit, from 4000 to 4001 Harvard Business machine-hours? What is the Publishing Education extra unit of capacity worth? Q.2 Assume that a second unit of engine assembly capacity is worth the same as the first. Verify that if the capacity were increased to 4100 machine hours, then increase in contribution would be 100 times that in Q.1. Q.3 How many units of engine 5|Page assembly capacity can be added before there is a change in the value of an additional unit of capacity? Learning Outcome: Linear programming: Managerial implications of linear programming in decision making 10 1. Students will be able to critically investigate the impact of changes in the objective function coefficient and the right-hand side of constraints on the optimal solution of a linear program, demonstrating their understanding of sensitivity analysis in linear programming Case: Dhebar A. (1989). Class discussion, lecture, and Merton Truck Company. numerical problems Harvard Business Case discussion questions: Publishing Education Q.1 Merton's production manager suggests purchasing Model 101 or Model 102 engines from an outside supplier in order to relieve the capacity problem in the engine assembly department. If Merton decides to pursue this alternative, it will be effectively “renting” capacity: furnishing the necessary materials and engine components and reimbursing the outside supplier for labor and overhead. Should the company adopt this alternative? If so, what is the maximum rent it should be willing to pay for a machinehour of engine assembly capacity? What is the maximum number of machinehours it should rent? Q.2 Merton is considering the introduction of a new truck, to be called Model 103. Each Model 103 truck would give a contribution of $2,000. The total engine assembly capacity would be sufficient to produce 5,000 Model 103s per month, 6|Page and the total metal stamping capacity would be sufficient to produce 4,000 Model 103s. The new truck would be assembled in the Model 101 assembly department, each Model 103 truck requiring only half as much time as a Model 101 truck (a) Should Merton produce Model 103 trucks? (b) How high would the contribution on each Model 103 truck have to be before it became worthwhile to produce the new model? Q.3 Engines can be assembled on overtime in the engine assembly department. Suppose production efficiencies do not change and 2,000 machinehours of engine assembly overtime capacity are available. Direct labor costs are higher by 50% for overtime production. While variable overhead would remain the same, monthly fixed overhead in the engine assembly department would increase by $0.75 million. Should Merton assemble engines on overtime? Q.4 Merton's president, in arguing that maximizing shortrun contribution was not necessarily good for the company in the long run, wanted to produce as many Model 101s as possible. After some discussion, it was agreed to maximize the monthly contribution as long as the number of Model 101 trucks produced was at least three times the number of Model 102s. What is the resulting "optimal" product mix? 7|Page Learning Outcome: 1. Students will demonstrate their ability to apply linear programming to propose solutions to a variety of managerial problems, utilizing their understanding of both linear programming and managerial decision-making processes Transportation and Pre-read: Chapter 17 Class discussion, lecture, and Assignment Problems: [Textbook, available online numerical problems Formulation and solution on book website] Case discussion questions: methodology MSIA: Optimizing 1. How will you convert this production, inventory and problem into a analytical distribution at The Kellogg problem? company 2. What will be the optimal plan so as to minimize the transportation costs? 11 3. What managerial insights can you derive from the optimal solution? Learning Outcome: Integer and Binary programming problems: Formulation and Solution methodology (branch and bound method) 1. Students will be capable of formulating a transportation problem as a linear programming (LP) model, and deriving the optimal solution. Pre-read: Chapter 7, pp. Class discussion, lecture, and 292 – 316 [Textbook] numerical problems Class discussion: 1. How to formulate integer program? 12 2. What is the need for additional integer constraint? 3. Can we not just round off the non-integer solution? Learning Outcome: Students will be able to formulate integer linear programming problem and 8|Page derive optimal solution. Case discussion questions: 1.How would this problem translate to optimization problem? 2. How should CEO allot the linac capacity over five years? Case discussion Case: Wang, J., Zaric, G.S. (2015). Radiation Treatment Machine Capacity Planning at Cancer Care Ontario. Ivey Publishing. Goal Programming Pre-read: Chapter 14 Class Discussion, lecture and some typical scenarios where [Textbook pp., 614 – 626] goal programming technique Chapter 15, pp 475 – 478 can be utilized. [Reference book-1] Learning Outcomes: 13 Development: Pre-emptive GP 14 Weighted Goal Programming through a Case Study Case: Chatterjee, D., Dhaigude, A. (2017). Apoorva: A Facility Location Dilemma. Ivey Publication. 1. Students will comprehend the process of resolving business problems with multiple goals, and derive best trade-off solution using technique of goal programming. Class Discussion, lecture and some typical scenarios where goal programming technique can be utilized. Case Discussion Questions: 1. What challenges is Rao facing? 2. What are the feasible locations for the new locations? 15 3. How should Rao evaluate the locations for his new restaurant? 4. Can Rao use the goal programming technique to select the best location? 5. Will the location change if Rao allocates weights to criteria based on his intuition about the weights recommended in the survey report? Analyze the different scenarios. 6. What should Rao do, and why? 9|Page Learning Outcomes: 16 Students will comprehend the concepts and techniques of weighted goal programming and will be equipped to make informed decisions. Decision Analysis: payoff Pre-read: Chapter 13, pp. Class discussion, lecture, and matrix, numerical, 544 – 550 [Textbook] numerical problems environments of decision Learning Outcome: making, decision making without probabilities: 1. Students will be able to Maximax (optimistic) apply varied decision-making criterion, maximin techniques to a given payoff (pessimistic or matrix and evaluate the conservative criterion), potential outcomes in order to Hurwicz criterion, laplace criterion, minimax regret suggest the best alternative criterion (savage criterion) solution. Decision Analysis: Pre-read: Chapter 13, pp. Introduction to decision 550 – 558 [Textbook] tree, decision making under risk (With probability), EVPI, EVwPI, EVwoPI, risk and sensitivity analysis, expected opportunity loss. 17 18 Class discussion, lecture, and numerical problems Class discussion: 1. How is decision making under risk different? Learning Outcome: 1. Students will demonstrate the ability to suggest the best course of action under conditions of uncertainty, applying their understanding of probability theory when the likelihood of each state of nature is provided. 2. Students will be capable of utilizing Expected Value of Perfect Information (EVPI), Expected Value with Perfect Information (EVwPI), and Expected Value without Perfect Information (EVwoPI) to determine the most suitable operational decision, Decision Analysis: Pre-read: Chapter 13, pp. Class discussion, lecture, and Decision analysis using 559 – 572 [Textbook] numerical problems sample information (Bayesian analysis), Class discussion: expected value of sample 1. Concept of Decision node information, risk profile, 10 | P a g e computing branch probabilities with Bayes’ theorem, EVSI, EVwSI, EVwoSI, efficiency of sample information and chance / outcome node. 2. What is the benefit of Bayesian approach? Learning Outcome: 1. Students will demonstrate the ability to suggest the best course of action under conditions of uncertainty. 2. Students will be capable of utilizing Expected Value of Sample Information (EVSI), Expected Value with Sample Information (EVwSI), and Expected Value without Sample Information (EVwoSI) to determine the most suitable decision strategy, applying these concepts to analyze and compare potential outcomes. 3. Students will be able to determine the efficiency of sample information, analyzing its impact on decision-making and outcome predictions. 19 20 Reading List and References Group Presentations Group Presentations Project Project Textbook: Anderson et al. (2019). An introduction to management science. 15th ed. Cengage Learning (must be comprehensive Reference books: and complete 1. Kumar, U. D. (2022). Business analytics. 2nd ed. Wiley. with all 2. Kreating et al. (2022). Forecasting and predictive analytics. McGraw Hill. details.) Prepared by Faculty Team Area & Program chairpersons Dr. Abhinav Sharma (Course Anchor) 11 | P a g e Approved by Associate Deans Approved by Dean SBM Sticker for date of receipt and attachments rubric and project guidelines MOBD Project Guidelines (AY 2024 – 2025) As a part of MOBD coursework, students are required to do a group project which carries 25 marks towards final assessment. Each group must identify a business domain and apply the techniques discussed in class to solve problem You can take primary / secondary literature based projects also. For e.g. refer to literature and understand how a particular organization/ industry has benefited from using optimization? Further, you should refer to research papers, web articles, cases for project work. The final project should cover Identification of Issues, Selection of Methodology, Analysis, Strategic Recommendations and Performance Implications. Project proposals must be submitted by 10th session EOD. You are advised to remain in touch with faculty for feedback and guidance. Report submission is due on 18th session of the course and presentations + viva will be conducted in 19th and 20th session. Project Report Guidelines 1. The report must be written in a word processing software. 2. First page should consist of tile of project, name, and roll number of group members. 3. Use a spacing of 1.15 and text must be justified across margins. 4. Use a font size of 12 (headings + paragraphs). 5. Headings and sub-headings must be numbered, bold and left aligned. 6. Use multi-level list to number headings and sub-headings. 7. Table title should appear on top of table along with table number. 8. Figure caption should appear on bottom along with figure number. 9. Your project report should contain (not limited to): (a) Introduction and identification of issues (b) Methodology (c) Analysis (d) Results (e) Strategic recommendations and performance implications 12 | P a g e (f)Conclusions 10. References should be cited in APA 7th edition format. The project rubric is given on next page. 13 | P a g e Course: Modelling and Optimization for Business Decisions Trimester II Assessment: Group Project Evaluation Faculty : Division : Group No. : Criteria Performance CLO 1: Process View and Mathematical Modelling Exemplary [9-10] The project demonstrates a deep understanding of the underlying processes in the business situation and uses mathematical models to analyze them. Good [6-8] The project demonstrates a solid understanding of the underlying processes in the business situation and uses appropriate mathematical models to analyze them. Average [4-5] The project demonstrates some understanding of the underlying processes in the business situation and uses basic mathematical models to analyze them. Poor [<4] The project does not demonstrate a clear understanding of the underlying processes in the business situation or does not use mathematical models appropriately to analyze them. CLO 2: Analytical Skills Exemplary [9-10] The project uses appropriate tools and techniques to solve complex business problems, demonstrating a high level of proficiency in quantitative analysis. Good [6-8] The project uses appropriate analytical tools and techniques to solve business problems, demonstrating a solid level of proficiency in quantitative analysis. Average [4-5] The project uses some analytical tools and techniques to solve business problems, but may lack depth or sophistication in quantitative analysis. Poor [<4] The project does not use appropriate analytical tools and techniques to solve business problems or shows a lack of understanding of quantitative analysis. Exemplary [5] The project evaluates a range of alternatives and uses objective criteria to choose the best option, demonstrating a high level of proficiency in decision-making. Good [4] The project evaluates alternatives and uses appropriate criteria to choose the best option, demonstrating a solid level of proficiency in decision-making. Average [2] The project evaluates some alternatives but may not use appropriate criteria to choose the best option, or the decisionmaking process may lack objectivity. Poor [1] The project does not evaluate alternatives or does not use appropriate criteria to choose the best option, or the decision-making process lacks objectivity. CLO 3: Evaluation and Decision-making Marks Obtained Total Signature of the Faculty Date 14 | P a g e 15 | P a g e