AN IMPROVED GENETIC ALGORITHM FOR JSP WITH FUZZY PROCESSING TIME AND DUE-DATE

摘要

Genetic algorithm （GA)for Job-Shop scheduling problem （JSP)with fuzzy processing time and due-date has been getting an increasing attention recently. But slow convergence and low precision still exist in these applications. To improve the performance of the existed GA for JSP with fuzzy processing times and due-date, a new approach（KRGA)is presented in this paper. A new crossover operator is designed, named as keeping-segments-reverse-crossover, which aims to make full use of the excellent quality of the parent chromosomes. The designed new operator can ensure validity and diversification of the evolving descendants, and can improve the accuracy of the optimal solution meanwhile. Through the numerical example of 10×10 Sakawa problem, compared with standard G&T and improved G&T, the feasibility and effectiveness of KRGA are illustrated.