本論文主要在探討當工作可分割時之等效平行機台排程問題。在本問題中,假設一個工作可以被分成許多的子工作,並且可將這些子工作指派到不同的機台上同時加工。每個工作具有已知的加工時間、準備時間以及交期。在機台加工過程中,當欲加工工作與前一加工工作不同時,必須耗費一相依整備時間方能進行加工。為最小化此問題之總延遲時間,本論文提出一個兩階段式方法來求解此問題。第一階段中,本論文提出五種初步排程產生方法。接著在第二階段中,將在第一階段產生的排程中有延遲的工作或子工作做進一步分割,並將分割後的子工作重新指派到機台上以減少總延遲時間。由實驗結果顯示,本論提出之方法表現優於前人所提出之方法。
This study focuses on the problem of scheduling jobs on identical parallel machines with job-splitting property. In this problem, it assumed that a job can be split into several sub-jobs, and these sub-jobs can be processed on different machines simultaneously. Each job has a processing time, a ready date and a due date. If a sub-job of a job is assigned after a sub-job of another job on a machine, a sequence-dependent setup time is incurred. A two-phase heuristic algorithm is proposed to minimize the total tardiness. In phase 1, five efficient heuristic methods are proposed. In phase 2, the sub-job of tardy jobs in the schedule provided by phase 1 are further split, and the split sub-jobs are rescheduled on the machines to reduce the total tardiness. Computer experiments show that the proposed algorithm outperforms a previously suggested algorithm.