Hybrid strategy of multi-objective differential evolution (H-MODE) for multi-objective optimisation
by Ashish M. Gujarathi; B.V. Babu
International Journal of Computational Intelligence Studies (IJCISTUDIES), Vol. 2, No. 2, 2013

Abstract: Evolutionary multi-objective optimisation (EMO) algorithms are preferred for solving the multi-objective optimisation (MOO) problems due to their ability of producing multiple solutions in a single run. In this study, hybridisation of the traditional sequential simplex method is considered with the evolutionary multi-objective differential evolution (MODE) algorithm for solving MOO problems. The hybrid strategy of MODE ensured that both the speed and accuracy are attained in a single algorithm. Various strategies of MODE algorithm are tested on several benchmark MOO test problems [both constrained (namely, SCH, FON, KUR, ZDT1, ZDT2, ZDT4, and ZDT3 and ZDT4) and unconstrained (namely, CONSTR and TNK)]. Two widely accepted performance metrics (convergence and diversity) from the point of view of MOO study are considered for evaluating the performance of strategies of MODE algorithm. Pareto fronts are obtained using newly developed strategies of MODE and are compared with the Pareto front obtained using other EMO strategy (NSGA-II). It is found that all the developed strategies of MODE algorithm converge to the true Pareto front for most of the test problems. However, the strategies of MODE result in slightly lower value of diversity metric as compared to NSGA-II for most of the test problems considered in this study.

Online publication date: Sat, 19-Jul-2014

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Computational Intelligence Studies (IJCISTUDIES):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com