A Genetic Algorithm for Multiobjective Optimisation using the interactive Sequential Multiobjective Problem Solving method

摘要

Despite the number of approaches established for Multiple Objective Optimisation Problems, few of them have been developed for the decision making process. This paper describes an interactive method to handle the Decision-Maker's preferences using a Genetic Algorithm. The Genetic Algorithm uses the Sequential Multiobjective Problem Solving method (SEMOPS) to describe the fitness function based on goal levels and aspiration levels. The aspiration levels are directly related to the Decision-Maker's preferences. Using real-valued representation the algorithm is applied to a case study for the problem of pollution in an artificial river basin. The results of this algorithm are compared with those found using the cutting-plane programming technique.