This paper introduces the notion of using co-evolution to adapt the penalty
factors of a fitness function incorporated in a genetic algorithm (GA) for
numerical optimization. The proposed approach produces solutions even bett
er than those previously reported in the literature for other (GA-based and
mathematical programming) techniques that have been particularly fine-tune
d using a normally lengthy trial and error process to solve a certain probl
em or set of problems. The present technique is also easy to implement and
suitable for parallelization, which is a necessary further step to improve
its current performance. (C) 2000 Elsevier Science B.V. All rights reserved
.