Investigations into the master timetabling problem

Abstract

Timetabling is a varied and interesting topic in the field of operational research. The problem of scheduling courses for departments in a university environment, like the University of Calgary, is no exception. Two methods are examined in order to try to solve this problem. First, a tailor made heuristic is developed. Next, a genetic algorithm, a more generic problem solving metaheuristic, is also developed and applied towards this problem. The Biology/Chemistry department's courses are used as a challenging test bed for these two approaches and the results are compared with the actual schedule used in the 1994-1995 academic year. Both methods could create schedules better than the actual schedule with the heuristic creating the better solutions between the two methods.