A game model based co-evolutionary for constrained multiobjective optimization problems

GaoPing Wang,Yongji Wang
DOI: https://doi.org/10.1109/ISCIT.2005.1566828
2005-01-01
Abstract:The use of evolutionary algorithms (EAs) to solve problems with multiple objectives (known as Multiobjective Optimization Problems (MOPs)) has attracted much attention recently. Population based approaches,such as EAs, offer a means to find a group of pareto-optimal solutions in a single run. However, most studies are undertaken on unconstrained MOPs. Recently, we developed the co-evolutionary algorithms for unconstrained MOPs. The objective of this paper is to introduce a modification to co-evolutionary algorithms for handling constraints. The solutions, provided by the proposed algorithm for one test problem, are promising when compared with an existing well-known algorithm. © 2005 IEEE.
What problem does this paper attempt to address?