A Hybrid Exam Scheduling Technique Based On Graph Coloring And Genetic Algorithms Targeted Towards Student Comfort

Scheduling is one of the vital activities needed in various aspects of life. It is also a key factor in generating exam schedules for academic institutions. In this paper we propose an exam scheduling technique that combines graph coloring and genetic algorithms. On one hand, graph coloring is used to order sections such that sections that are difficult to schedule comes first and accordingly scheduled first which helps in increasing the probability of generating valid schedules. On the other hand, we use genetic algorithms to search more effectively for more optimized schedules within the large search space. We propose a two-stage fitness function that is targeted toward increasing student comfort. We also investigate the effect and potency of the crossover operator and the mutation operator. Our experiments are conducted on a realistic dataset and the results show that a mutation only hybrid approach has a low cost and converges faster toward more optimized schedules.