Journal of Xidian University ›› 2016, Vol. 43 ›› Issue (3): 49-54+130.doi: 10.3969/j.issn.1001-2400.2016.03.009

Previous Articles     Next Articles

Improved NSGA-Ⅱ algorithm based on the uniformly crowding distance

WANG Mingzhao;WANG Yuping;WANG Xiaoli;WEI Zhen   

  1. (School of Computer Science and Technology, Xidian Univ., Xi'an  710071, China)
  • Received:2015-03-10 Online:2016-06-20 Published:2016-07-16
  • Contact: WANG Mingzhao E-mail:


With the wide application and further study of the genetic algorithm in multi-objective optimization problems, the NSGA-Ⅱ has been one of the representative evolutionary algorithms for multi-objective optimization problems. Crowding distance in the NSGA-Ⅱ plays an important role in convergence and uniform distribution of the solutions, but the NSGA-Ⅱ does not fully take the effect of each individual and the whole population into consideration. To estimate the region density more reasonably so as to make the solution set more uniformly converge to the Pareto optimal front, we design a uniformly crowding distance operator based on the uniformly crowding range and Gini weight, and propose an improved NSGA-Ⅱ algorithm. Finally, the effectiveness of the proposed algorithm is verified by experiments on six multi-objective optimization test functions.

Key words: uniformly crowding range, Gini weight, uniformly crowding distance, multi-objective optimization