Multi-objective problem, multi-species solution: An application of the cellular genetic algorithm

Multi-objective optimisation problems typically consist of distinct objectives that cannot be optimised simultaneously. The goal is to find a set of good solutions - the Pareto-optimal set - where each solution is not dominated by any other solution. In this paper we present a novel approach for approximating the Pareto-optimal set by using the Cellular Genetic Algorithm (CGA). The CGA maps the evolving population of solutions onto a pseudo landscape. The introduction of multiple species, one for each objective to be optimised, combined with the dynamic spatial structure of the CGA allows solution diversity to be maintained. We investigate the performance of the algorithm using a variety of test problems. Preliminary results indicate that a wide range of non-dominated solution can be found.