An Evolutionary Immune Approach for University Course Timetabling

The university course timetabling problem (UCTP) is a combinatorial NP-complete problem that has been subject to research since the early 1960’s. Numerous solution techniques have been applied to the timetabling problem ever since. This paper aims at formulating an immune-inspired algorithm, namely the Clonal Selection Algorithm1 (CSA1) and testing its ability in solving the UCTP against the Genetic algorithm (GA). An Immune-Genetic algorithm (IGA) was also created, which combines the crossover operator borrowed from the genetic algorithm with immune-inspired concepts. Also, experimenting with the effects of changing the selection and re-selection schemes of the algorithms motivated the creation of a second version of CSA1, that is CSA2 and three more versions of IGA: IGA1, IGA2 and IGA3. All the devised algorithms were contrasted in their performances against the GA. Enhancements were applied to the mutation operator of the formulated algorithms by introducing a ‘move factor’. As a means of improving the results attained by the algorithms, local search consisting of three variable neighborhoods was incorporated into each of them. The algorithms were tested over two problem instances, with varying complexities and the results demonstrate the effectiveness of the algorithms in solving the UCTP.