HIGH SCHOOL TIMETABLING USING TABU SEARCH AND PARTIAL FEASIBILITY PRESERVING GENETIC ALGORITHM

The high school timetabling is a combinatorial optimization problem. It is proved to be NP-hard and has several hard and soft constraints. A given set of events, class-teacher meetings and resources are assigned to the limited space and time under hard constraints which are strictly followed and soft constraints which are satisfied as far as possible. The feasibility of timetable is determined by hard constraints and the soft constraints determine its quality. Difficult combinatorial optimization problems are frequently solved using Genetic Algorithm (GA). We propose Partial Feasibility Preserving Genetic Algorithm (PFP-GA) combined with tabu search to solve hdtt4, “hard timetabling” problem a test data set in OR-Library. The solution to this problem is zero clashes and maintaining teacher's workload on each class in given venue. The modified GA procedures are written for intelligent operators and repair. The PFP-GA in association with Tabu Search (TS) converges faster and gives solution within a few seconds. The results are compared to that of using different methodologies on same data set.