 |
    |
Autonomous Agent for Multi-objective Optimization
Alain Berro and Stephane Sanchez
IRIT-UT1, Université Toulouse I Sciences Sociales, 1 place Anatole France 31042 Toulouse Cedex, Franceberro@univ-tlse1.fr
sanchez@univ-tlse1.fr
Abstract. In this article, we present an agent-based method associated with a local research inspired by strategies of evolution to solve multiobjective problems. In comparison with GA-based methods this method uses few parameters. Moreover a decision maker can easily understand the influence of these parameters on the result. The conception of this method led us to represent the Pareto optimal set with zones and not with points. This representation gives additional information which allows to choose between two non-dominated solutions.
LNCS 3102, p. 251 f.
lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2004