In this paper we propose a strategy to incorporate heuristic knowledge into the initial population of a Genetic Algorithm to solve Job Shop Scheduling problems. This is a generalization of strategy we proposed in a previous work. The experimental results reported confirm that the new strategy improves the former one. In particular, a higher diversity in achieved among the heuristic individuals, and at the same time the mean fitness is improved. Moreover, these improvements translate into a better convergence of the GA. © Springer-Verlag Berlin Heidelberg 2001.
CITATION STYLE
Vela, C. R., Varela, R., & Puente, J. (2001). Initialization in genetic algorithms for constraint satisfaction problems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2084 LNCS, pp. 693–700). Springer Verlag. https://doi.org/10.1007/3-540-45720-8_83
Mendeley helps you to discover research relevant for your work.