125. Constraint Programming formulation for a real-world final exam scheduling problem with parallel sessions based on short time intervals
Invited abstract in session WC-3: Mixed Integer Programming I, stream Discrete Optimization.
Wednesday, 10:00 - 11:30Room: C 104
Authors (first author is the speaker)
| 1. | László Kálmán Trautsch
|
| Department of Automation and Applied Informatics, Budapest University of Technology and Economics | |
| 2. | Bence Kovari
|
| Department of Automation and Applied Informatics, Budapest University of Technology and Economics |
Abstract
Scheduling final exams is subject to various requirements that differ by countries and universities. A range of personal, institutional, and regulatory factors should be considered at the same time for creating an optimal schedule. We propose a Constraint Programming model for scheduling final examinations at the Department of Automation and Applied Informatics, Budapest University of Technology and Economics. The requirements of this scheduling problem regard an examination period which is divided into 5-minute intervals. Heterogeneous student groups and instructors with various roles are scheduled to parallel sessions based on these time intervals. The cost function is defined by multiple types of constraints, such as balancing the workload, optimizing start and end times of sessions and breaks, and penalizing leaks in the schedules of instructors. We present the formulation of the various complex requirements of final exam scheduling and demonstrate the results of applying a Constraint Programming solver on our model to find feasible solutions for a real-world scheduling problem involving 101 students.
Keywords
- Mixed integer nonlinear optimization
Status: accepted
Back to the list of papers