Solving The Examination Timetabling Problem Using A Hybrid Of Particle Swarm Optimization And Grey Wolf Optimizer

Abstract

Constructing an examination timetable in University can be defined as a challenging task with many practical constraints. There are various conditions to seize a good feasible timetable. In this work, we want to solve the examination timetabling problems for University in Vietnam, acknowledged that most Universities in Vietnam have limited facility. This shortage leads to a lack of sufficient resources such as classrooms and laboratory sessions. Generally speaking, it is very important to schedule the examination while maintaining that the resources are used efficiently. This work proposed an exam meta-heuristic to solve the examination problem using a hybrid of Particle Swarm Optimization (PSO) and Grey Wolf Optimizer (GWO) to find the optimal solution while ensure to bring out a satisfying result. Notably, this method has never been applied to compete for the best-known results in the field of examination timetabling. So we want to adopt the approach to investigate its efficiency. We then compared the result in small cases with IBMCPLEX solutions to validate our metaheuristic. To enhance the result, we have added a multi-search method to improve the searching ability of the proposed approach. Our approach provides a reasonable result with a large scale study in International University consisting of more than 400 exams while CPLEX was not able to build a proper solution due to the large size problem.