M.Sc. Tezi Görüntüleme

      The University Course Timetabling Problem is one of the most famous scheduling problems, which has attracted many researchers to solve the problem. This problem is located in NP-Hard problems group and so deterministic methods are not efficient to solve it and heuristic and meta-heuristic methods are suggested to do so One of the meta-heuristic methods suggested to solve optimization problems in continuous environments that is inspired by the swarm intelligence of honey bees in search for the food, is called ABC algorithm. This algorithm could achieve good results compared to other meta-heuristic methods in a continuous environment. The task of adapting ABC to solve problem in discrete environments, e.g. University Course Timetabling, needs defining the neighborhood concept and applying this concept to find the better solutions. In this thesis, ABC has been used to solve one of popular types of University Course Time-Tabling problems which is Curriculum Based University Course (CB-CTT) Time-Tabling problem. It was the third type of Time-Tabling problems in ITC-2007 competition. The adaption of ABC to solve this problem gets the advantage of some improvements in neighbor generation and using local search technique, compared to the original ABC algorithm.

      Having a benchmark case study for this problem helps researchers to work on a standard dataset and makes it easy to compare different results obtained by different researchers with different approaches. Our achieved results on 21 datasets of ITC-2007 is comparable to five winners of this competition.

      Key Words: Meta-heuristic method, Artificial bee colony, University course timetabling