Abstract: Considering that evolutionary multiobjective optimization has been becoming one of research topics in evolutionary algorithm community recently， a speciesbased mechanism was introduced into multiobjective GA， which is initially applied into GA for multimodal optimization problem. And then a new speciesbased method was designed and proposed for multiobjective optimization problem， which comprises a species seed indentifying strategy based on Pareto optimality and crowd degree and an adaptive species constructing scheme. Experimental results showed that the proposed speciesbased multiobjective GA outperforms nondominated sorting GA Ⅱ(NSGAⅡ)on a set of benchmark test problems.

Key words: multiobjective optimization problem, evolutionary multiobjective optimization, genetic algorithm, speciesbased mechanism, multimodal optimization