抽象的

A New Ranking Scheme for Multi and Single Objective Problems

A R Khaparde , V M Athawale

Evolutionary algorithms (EAs) have received a lot of interest in last two decade due to the ease of handling the multiple objectives. But one of criticism of EAs is lack of efficient and robust generic method to handle the constraints. One way to handle the constraint is use the penalty method .in this paper we have proposed a method to find the objective when the decision maker (DM) has to achieve the certain goal. The method is variant of multi objective real coded Genetic algorithm inspired by penalty approach. It is evaluated on eleven different single and multi objective problems found in literature. The result show that the proposed method perform well in terms of efficiency, and it is robust for a majority of test problem