Pareto-based Cost Simulated Annealing for Multiobjective Optimization

In this paper, a multiobjective simulated annealing (MOSA) method is introduced and discussed with the multiobjective evolutionary algorithms (MOEAs). Though the simulated annealing is a very powerful search algorithm and has shown good results in various singleobjective optimization fields, it has been seldom used for the multiobjective optimization because it conventionally uses only one search agent, which is inadequate in finding many solutions of the Pareto set. With the idea that the simulated annealing has a uniform state probability over global optima, a new multiobjective simulated annealing method is suggested. The experimental performance of the developed algorithm is compared with multiobjective evolutionary algorithms and shows that the proposed simulated annealing has good uniformity properties.