Skip to Main Content
Niche genetic algorithm (NGA) is superior to genetic algorithm (GA) in multiple hump function optimization. NGA could search all global optimums of multiple hump function in a running. It is a class of parallel evolutionary method which suppresses genetic drift by forming stable subpopulations to maintain population diversity. To algorithm population diversity plays an important role to avoid trapping in premature convergence. In this paper diversity methods and measures from the literature are introduced. An obvious contrast between NGA and GA has been analyzed in diversity. The results show that NGA is of good advantage to maintain diversity during different stage of the evolutionary process.