An Approach For Multiobjective Optimization: Pareto-MEC

In the real world, there are many multi-objective optimization problems. The multi-objective optimization is a rising subject in the recent 30 years.In this paper a new approach is proposed. It introduces the conception of Pareto into the Mind Evolutionary Computation (MEC) which was first proposed by Chengyi Sun in 1998. MEC imitates two phenomena of human society - similartaxis and dissimilation and overcomes the problems of GA successfully.The principles of Pareto-MEC are: (1) A number of individuals are scattered in the whole solution space, and then some better individuals of them are selected as the initial centers for every group. (2) Each group only searches a local area and gradually shifts from its initial center to the Pareto front. (3) During the process of shift to this front, this algorithm would bound the searching region of the group and control the shifting direction of the group. Both of above function (1) and function (3) are called as dissimilation, and function (2) is called as similartaxis.Pareto-MEC is respectively compared with the reference algorithms of Rand, VEGA, NSGA and SPEA. The test functions used in the experiment are a suit of four different test problems: convexity, non-convexity, discreteness and non-uniformity. On alltest problems, Pareto-MEC outperforms Rand, VEGA and NSGA; Pareto-MEC is as good as SPEA on the first three test problems; yet it beats SPEA on the last test problem. Different from the reference algorithms that use the pre-specified generation number as their terminations, Pareto-MEC has an objective termination criterion that can ensure the quality of solutions and the computational efficiency.Two evaluative methods: Cover and Spacing are used as the quantificational criterion for our algorithms on the test functions: convexity, non-convexity and non-uniformity. The experiment shows that the performance of Pareto-MEC is as good as SPEA which is one of the excellence algorithms in the world.This is the first time that using MEC resolving multi-objective optimization problems. And the MEC had never used the pareto conception before. The experiment results of Pareto-MEC are as good as that of SPEA. Yet Pareto-MEC has less number of generations. Moreover, this approach has an impersonal stopping condition not as other methods mentions here which use a given iteration times as their teminatind conditions. It can be seen that Pareto-MEC is very suit multi-objective optimization problems.