| Summary: | 碩士 === 國立臺灣科技大學 === 工業管理系 === 95 === A no-wait scheduling problem indicates a circumstance where there are no allowed buffers within each operation from job release to finish with the technological constraint. Common examples of a no-wait restriction are the production line in semiconductor fabrication facilities, steel factories, chemical process, and so on. In order to meet the constraint, jobs in queue have to wait for a span of time to avoid possible collision with discarding. Thus, the delay time intervals between jobs must be predetermined. Production with this constraint often goes along with batch processing to make up a deficiency of inefficient throughput due to longer processing time. This thesis proposes an approach incorporating a heuristic algorithm and a genetic algorithm to deal with the determination of accurate delay time intervals and efficient dispatching rule for the scheduling problem with the objective of minimizing maximum completion time (makespan). Furthermore, an analysis of a wet etching station in the semiconductor fabrication facilities will be presented to demonstrate the traditional method and the proposed approach. The computational results show that the proposed approach can improve the solution performance.
|
|---|
