S1440102 遺伝的アルゴリズムによるフレキシブルジョプショツプスケジューリングの一手法

Abstract

In these two decades, high-mix low-volume manufacturing and agile manufacturing have been the mainstream. This made discussion about job shop scheduling more active, and many methods based on metaheuristics, especially genetic algorithm (GA), have been proposed. However, many actual factories have alternative machines, and scheduling problem for job shops with alternative machines, i.e. flexible job shop scheduling problem (FJSP) has been discussed recently. To solve FJSP, it is necessary to perform machine selection and job selection. In conventional methods using GA, these two decision making are carried out sequentially. This would fail to search a large solution space and to obtain a good solution. In this paper, we propose a new scheduling method using GA in which the machine selection and job selection are integrated by using random key coding.