| Study program / study programs: Advanced data analytics |
| Course: Analytics and оptimization |
| Teacher(s): Milan Stanojević, Dragana Makajić-Nikolić, Gordana Savić, Marija Kuzmanović |
| Course status: Elective |
| ECTS points: 10 |
| Course objective: Introduction to analytics and optimization with aim of optimal decision making using quantitative models and methods. |
| Learning outcomes: Students will be able to identify and analyze real world problems and data collected in the process. Students will be able to formulate real world problem as an optimization problem. Students will be able to use optimization methods and techniques (especially methods and techniques for solving linear and integer programming models). Students will be able to analyze and visualize results. |
| Course structure and content: Basics of analytics and role of optimization in analytics. Descriptive analytics (basics of data clearing, missing data handling, summarizing, grouping, classifying, clustering, visualizing,…). Predictive analytics (basic of forecasting using trend, smoothing, regression functions…). Prescriptive analytics – mathematical modelling and optimization. Methods for solving mathematical models. Sensitivity analysis. Multicriteria analysis. Heuristics. Methods for alternative selection in the presence of uncertainly. Application of all methods using MS Excel tools, AMPL… Solving real life problems and presenting the results. |
| Literature/Readings: J.A. Lawrence, B.A. Pasternack, Applied Management Science, John Wiley & Sons Inc. 2002. A. Makhorin, Modeling Language GNU MathProg Language Reference, Free Software Foundation, 2013. R. Saxena, A. Srinivasan, Business Analytics: A Practitioner’s Guide, Springer, 2013 J. R. Evans, Business Analytics: Methods, Models and Decisions, Pearson, 2013 R. Fourer, D.M. Gay, B.W. Kernighan, AMPL: A Modeling Language for Mathematical Programming, Duxbury Press / Brooks /Cole Publishing Company, 2002. |
| The number of class hours per week: Lectures: 4 Labs: 0 Workshops: 0 Research study: 3 Other classes: 0 |
| Teaching methods: Lectures are followed by the corresponding presentations; all models will be illustrated in the hypothetical example. Students will, through case studies using appropriate software, analyze the input data, define quantitative models, analyze results, and make alternative scenarios with the aim to set the basis for decision making. |
| Evaluation/Grading (maximum 100 points): Pre-exam requirements: 70 (Participation in class – 10, Case study – 60) Final exam (Written exam): 30 |
