A cooperative system for metaheuristics and its application

Optimization problems is defined as function required finding optimum state depends on parameters, which are certain limitations. In field of optimization, aim is to find the optimal solution or approximate solution from many alternative solution that provides all the restrictions. Metaheuristic is an extremely effective method for finding approximate solutions to optimization problems. However, when metaheuristics used, algorithm selection problem arises. This problem is to decide that studied optimization problem is solved by in which algorithm so as to maximize performance measure. In this thesis, purpose is to use a cooperative system for combine different metaheuristics brilliantly to deal with algorithm selection problem. An intelligent combination of different metaheuristics is expected to provide more flexible, more efficient and more robust approaches. But it requires the less precision in order to obtain more robustness. This situation is generated based on a methodology designed with soft computing. In addition algorithm selection problem, adjusting algorithm parameters have significant importance to obtain good results. For this reason, proposed cooperative system give a fine tuning of parameters depends on soft computing techniques. In order to test usefulness of this approach, comparison is made between results obtained by proposed system and individual metaheuristics.