A fuzzy-based multi-term genetic algorithm for reentrant flow shop scheduling problem

In semiconductor manufacturing factories, the process of wafer fabrication is the most technologically complex and capital intensive stage. This process is configured as a reentrant flow shop process with many machines and processing steps. It needs an efficient and effective scheduling method for large size process in order to increase the competitiveness. The reentrant flow shop problem (RFSP) means that all jobs have the same route through the shop machines and the same shop machine is used several times to complete a job. This research provides an effective fuzzy-based multi-term genetic algorithm to solving RFSP with the objective of minimizing the total turn around time (TTAT). The proposed method focuses on the critical point in scheduled solutions. The middle position of longest TAT is defined as the critical point. According to the critical point and current generation, fuzzy logic chooses the focused term of chromosome, then the genetic algorithm effects on this term. In each evolution, only corresponded part of chromosome is evolved by crossover and mutation while other parts of chromosome remain unchanged. Through computational experiments, the effectiveness of the fuzzy-based multi-term genetic algorithm is evaluated.