Abstract: NSGA-II is a multi-objective evolutionary algorithm，and its performance is so good that it has become very popular in the last few years.To improve its bi-objective optimization efficiency，in this paper，a layering strategy according to need is adopted and so a new algorithm to construct the set of non-dominated fronts is proposed to replace the original method of NSGA-II.Compared with the NSGA-II’s computational complexity（O（N²）），the new algorithm’s computational complexity is reduced to O（kN＋NlogN），k is the number of fronts，and k<<N.The experiment results also show that there are fewer layers of non-dominated fronts，counts of dominate compare and much less running-time in the new approach compared with NSGA-II.

Key words: multi-objective evolution, non-dominated front, layering strategy according to need

摘要： NSGA-II是一种性能优良的多目标进化算法，近年来非常流行。为了进一步改进NSGA-II在双目标优化时的效率，采取了按需分层的策略，提出了一种新的非支配前沿集分层方法以替代NSGA-II原有的分层方法。与NSGA-II的时间复杂度O（N²）相比，新方法的时间复杂度减少为O（kN＋NlogN），k为所分前沿层数（k<<N）。实验结果也表明，新方法与NSGA-II相比具有更少的非支配前沿层数，支配比较次数和运行时间。

关键词: 多目标进化, 非支配前沿, 按需分层