No abstract

No abstract

This paper presents a coevolutionary approach to the numerical optimization of large facility layouts. Our work is based on a mixed integer model for the layout constraints and objectives, which improves formulations found in the literature. Nevertheless, layouts with more than seven departments are difficult to solve. One way out is to apply genetic algorithms - searching systematically for solutions but without guarantee of finding an optimum. In this paper we suggest some improved mutation and cross-over operators. Yet, with increasing number of departments also genetic algorithms take very long. In this case we propose to use additional structures given by qualitative or quantitative reasoning. Clustering the departments into groups we allow each group ('species') to evolve (genetic algorithm) in a separate area while position and size of these areas ('environment') undergo an evolution, too. Numerical experiments verify this coevolutionary approach

No abstract

No abstract