Cluster based solution exploration strategy for multiobjective particle swarm optimization

This paper introduces the solution exploration strategy into particle swarm optimization (PSO) to distribute local guides for each particle of the population to lead them find out the solutions of Pareto optimal set. After solution found, we utilize cluster concept to sift representative nondominated solutions from the external repository to keep their diversity. We also incorporate a mutation like operator that enhances the solution searching capability. We compared our method to other related MO methods. These methods are examined on different test functions and the results are compared with the results of multiobjective evolutionary algorithm (MOEA).