Scheduling with Memetic Algorithms over the Spaces of Semi-active and Active Schedules
The Job Shop Scheduling Problem is a paradigm of Constraint Satisfaction Problems that has interested to researchers over the last decades. In this paper we confront this problem by means of a Genetic Algorithm that is hybridized with a local search method. The Genetic Algorithm searches over the space of active schedules, whereas the local search does it over the space of semi-active ones. We report results from an experimental study over a set of selected problem instances showing that this combination of search spaces is better than restricting both algorithms to search over the same space. Furthermore we compare with the well-known Genetic Algorithms proposed by D. Mattfeld and the Branch and Bound procedure proposed by P. Brucker and obtain competitive results.
KeywordsGenetic Algorithm Local Search Critical Path Constraint Satisfaction Problem Memetic Algorithm
Unable to display preview. Download preview PDF.
- 8.Mattfeld, D.C.: Evolutionary Search and the Job Shop. In: Investigations on Genetic Algorithms for Production Scheduling, Springer, Heidelberg (1995)Google Scholar
- 10.Taillard, E.D.: Parallel Taboo Search Techniques for the Job Shop Scheduling Problem. ORSA Jorunal on Computing 6, 108–117 (1993)Google Scholar
- 11.Vaessens, R.J.M., Aarts, E.H.L., Lenstra, J.K.: A Local Search Template. In: Mnner and Manderick, pp. 65–74 (1992)Google Scholar
- 14.Varela, R., Serrano, D., Sierra, M.: New Codification Schemas for Scheduling with Genetic Algorithms. LNCS 3562, Springer-Verlag 145, 11–20 (2003)Google Scholar
- 15.Yamada, T., Nakano, R.: Scheduling by Genetic Local Search with multi-step crossover. In: Fourth Int. Conf. On Parallel Problem Solving from Nature (PPSN IV), Berlin, Germany, pp. 960–969 (1996)Google Scholar