Title: Using mixed integer programming to win a cycling game
Authors: Belien, Jeroen
Goossens, Dries
Van Reeth, Daam
De Boeck, Liesje
Issue Date: Jan-2010
Host Document: Orbel 24: 24th annual conference of the Belgian Operations Research Society: Program and abstracts pages:65-66
Conference: ORBEL edition:24 location:Luik date:28-29 January
Abstract: This paper presents an application of optimization modeling to win a popular cycling game. The application involves real-life data of today's cyclists and challenges the students because of the competition aspect. Since the developed optimization model contains features of knapsack problems, multiperiod inventory problems and integer programming modeling, it is perfectly suited as a concluding case study in an undergraduate operations research/management science course. Moreover, the application also sharpens the understanding of the working of stock exchange markets and is, therefore, also interesting for finance courses. The application was originally developed for an MBA operations research course focusing on spreadsheet modeling skills, but it can also be used in courses that focus on algebraic modeling of optimization problems. Recently, (mixed) integer programming has gained in popularity. Indeed, advances in solution methods as well as computer hardware and software make solving such problems no longer an exhaustive challenge with respect to computing times. As a result, also the importance and relevance of (M)IP increased for solving real-life managerial-related problems. However, getting students really excited about (M)IP is not always an easy task. One way to overcome this is to let them experience how MIP can be used in applications taken from their own living environment. In this paper, we use an a posteriori game component of a popular cycling game (called the Gigabike game), involving real-life data of all current cyclists. Further on, we stimulate the students by adding a competitive aspect: the week before the solution is discussed in class acts as a real race among all students in finding the best solution for the game. An Excel sheet facilitates this search process. The game is very much suited for an undergraduate OR/MS course as a case concluding the (M)IP chapter. Indeed, the model can be gradually built up in three clearly distinctive phases, each phase covering an important class of (M)IP optimization methods. In the first phase, we consider the model as made up of independent knapsack problems (one for each period). The second phase introduces dependency between the periods, turning it into a multi-period model. Finally, the third phase adds a typical IP constraint in order to obtain the right final model. The models of all phases can be solved using Frontline's standard version of Excel Solver, as such making the models accessible to every student. The class experience with the game fosters our belief that using applications from the student's own world of interest are very valuable in getting the students excited about what they themselves call theoretical models (in this case, MIP models). The competitive component as introduction to the game, has impressively stimulated the student's interest which was clearly visible during the class. We hope that in this way the students learned about the power such models can provide and more importantly, that they start to get a feeling about the importance and relevance these models might have in real-life management applications.
Publication status: published
KU Leuven publication type: IMa
Appears in Collections:Research Center for Operations Management, Leuven
Research Centre for Globalization, Innovation and Competition, Campus Brussels (-)
Faculty of Economics and Business (FEB) - miscellaneous
Research Centre for Quantitative Business Processes, Campus Brussels (-)
Department of ECON-CEDON, Campus Brussels

Files in This Item:

There are no files associated with this item.


All items in Lirias are protected by copyright, with all rights reserved.