本論文主要研究多階段不等效平行機排程問題,針對所有工作在某些製程可選擇自製或外包機器加工的情況,在有交期時間的限制下,所產生的延遲懲罰成本與外包成本,將所此兩種成本總合視為總成本。而研究目標乃是最小化總成本。 首先,研究在某些製程方面,工作可選擇在自製或外包機器上加工,而在考慮各工作的延遲懲罰成本與具有不同外包情形下的成本下,如何對各工作加以排序,最後以在所有工作完成時的總成本最小化為績效指標。 由於本論文研究之問題屬於NP-Hard,基於求解效率,將提出啟發式演算法並建立數學規劃模式,以作為評估求解品質成效的基準。由實驗結果分析得知,第四個啟發式演算法能夠有效的求取排程順序解。
This paper addresses the multi-stage non-identical parallel machines scheduling problem. In some stages, there are two kinds of machines. One is out-sourcing and the other is self-manufactured. Jobs are processed by out-sourcing or self-manufactured in some stages. The performance measure is the minimization of total cost, including the out-sourcing cost and the weighted tardiness cost. This problem is known to be NP-hard, four heuristic algorithms are provided and an integer programming model is presented for benchmarking. Experimental results show that the proposed fourth heuristic algorithm is effective and efficient.