analysing the impact of niching in multi-objective evolutionary algorithm based on decompose (moea/d)

Recently evolutionary algorithms have been successfully used for solving multi-objective optimization problems. evolutionary algorithms are trying to simulate the nature’s behavior in finding the best solution for problems. multi-objective evolutionary algorithm based on decomposed (moea/d) is an evolutionary framework for estimating the pareto optimal points by less computational complexity. this framework has some setting parameters which seem to have important role on achieving fine solutions. neighboring is such parameter that the best tuning of its boundary leads to improve the performance of moea/d. since the island model is one of the easiest niching methods that can study the manner of evolution in niches, this paper presents a non-uniform island moea/d to investigate the effect of neighborhood size on the performance of the solutions in the objective space. the experimental results show that the non-uniform island moea/d outperforms the original moea/d on the test instances.