Examination Timetabling Problem: A Case Study

Background/Objectives: This paper presents a real-world examination-timetabling problem associated with Universiti Malaysia Terengganu (UMT) in Malaysia. We aim to develop a mathematical model that considers the main requirements to produce the examination timetable at UMT and attempts to optimize the assignment of exam into room and timeslot as preferred by the communities whilst satisfying the requirements mentioned. Methods/Analysis: The main requirements is modelled using Binary Integer Programming method and is validated using a self-develop dataset with two different preferences assignment and the computational results are reported and analyzed directly using the AIMMS Software with CPLEX 12.6.3 solver. Findings: The results shows that the newly developed models have successfully produce an examination timetable that completely solve all basic requirements addresses by the university and through the application of the models, we manage to maximize the preferences of the communities with the assignment of exam into the best preferred slots. Novelty/Improvement: These results will be used as reference for developing a more sophisticated examination-timetable model that will incorporate more demanding and challenging constraints as well as the preferences from community, which will be significantly better than those constructed using the existing manual system especially in terms of solving all requirements which cannot be easily done using the former system in UMT.