An improved density estimation method in NSGA2

Multi-Objective Evolutionary Algorithms (MOEAs) are efficient and widely accepted ways in solving Multi- Objective Problems (MOPs), and NSGA2 may be the most popular one. Diversity distribution is one of the major indexes to reflect the performance of MOEAs. The diversity maintenance strategy in NSGA2, is a density estimation method, called crowding-distance also, based on which, a new density estimation method is proposed in this paper. The new method has been tested with five test problems, and it performs better in diversity preservation.