Study of hybrid approaches used for university course timetable problem (UCTP)

Abstract

University course timetabling problem (UCTP) is important issue forever. Scheduling problems are considered and proved NP-Complete by different researchers. It is due to continuous changing of constraints to be fulfilled on rapidly changing data. Different hybrid state of art techniques and their use for university course timetable problem is investigated in this study. This paper also analyze occurrence of constraints and there ratio of similarity in recent research trend on university course timetabling problem. Constraints highly vary from department to department. This level of variance of constraints makes timetabling problem difficult to solve and NP-complete. In recent years, concept of hybridization of different methods increased. This study analyzed use of hybrid heuristic and meta-heuristic methods for university course timetabling problem. In this study, we categorize this hybridization into two main categories: local search hybridization with local search based approaches and population based hybridization with local search based approaches. It is observed that Population based methods (Genetic Algorithm (GA), Particle Swarm optimization (PSO) and Artificial Bee colony (ABC) etc.) are preferred in combination with local search (LS) based methods for university timetabling problem. Concept of hybridization of population based methods with local and other population based methods is adopted to eliminate demerits of both methods. Although, hybrid methods are difficult and also require more computational cost. Still hybridization is useful for finding better solutions.