Genetic multi-criteria approach to flexible line scheduling

Job scheduling in flexible production systems is a complex task even for simple cases. This paper deals with this problem using fuzzy set theory and genetic algorithms. Fuzzy techniques allow us to define global performance measures expressing different and often conflicting objectives of the production. Moreover, according to a fuzzy multi-criteria algorithm, we propose a methodology to combine various heuristics, with different weights, in a single dispatching criterion. A genetic optimization process selects the weights guaranteeing good system performances. A case study and some extensive simulations show the efficiency of the methodology. (C) 1998 Elsevier Science Inc. All rights reserved.