論文名稱：考慮固定週期批量之物料需求規劃(MRP)生產前置時間最佳化 頁數：52 校所組別：國立中央大學工業管理研究所 畢業時間及提要別：102學年度第一學期碩士論文摘要 研究生：董嘉衛 論文提要內容： 摘 要 製造業生產計劃與管制的工具之中最為廣泛使用為「物料需求規劃」（material requirements planning, MRP），而在運用此生產規劃系統時，計劃前置時間（planned lead time, PLT）則為其中非常關鍵的參數，它直接或間接影響最終演算的規劃結果。當計劃前置時間設置較長，雖然可使生產變動趨向平滑卻容易導致存貨使用率降低，而在製品的水準（work-in-process, WIP）也會隨著需求變異提高，自然增加不必要的生產成本。 面對實際環境中更為複雜的因素，還包含生產批量或固定週期等生產條件，考驗前置時間設置的方式與智慧。對生產者而言計劃前置時間決定生產製造的服務水準，但又同時牽動庫存水準。如何在固定生產週期的生產條件下使用物料需求規劃找出最佳的計劃前置時間，符合現實生產限制的考量又可獲得最大的經濟效益是本研究最重要的課題。 本研究針對組裝式產業（assembly industry）做實際研討並使用混合整數規劃（mixed-integer programming, MIP）建置數學模型，以物料需求規劃為模型設計基礎，建立「IBM ILOG CPLEX Optimization Studio」最佳化軟體的開發環境，導入產業實際數據資料進行實驗分析。研究模型以區隔計劃前置時間、實際發放工單時間與實際生產時間為基礎，特別考量實際物料本身特性以及生產條件，針對物料在固定生產週期條件下計算計劃前置時間之最佳值，幫助決策者在固定物料生產週期的生產環境下運用物料需求規劃系統制定較適切的生產規劃。; Abstract MRP system is the most widely used tools to develop production and control plans in Manufacturning industry. A key parameter of MRP system is the planned lead time (PLT), it impact the final algorithm results directly or indirectly. In the past, we usually set long lead time to smooth the production curve or reduce the waiting time but also lead to decrease inventory utilization. In the other hand, work-in-process (WIP) level will increase accompany by demand uncertainty. To face the actual enviroment which have more complex factors include batch production, fixed period and other consideration, examed the manager how to set the lead time and use in prodution. For the producer ,the production service level determined by lead time and also control the inventory level. How to use the MRP system to find the optimal solutions of planned lead time in fixed period production condition and also fit the actual production restrict to make the profit maximization is the key point in this paper. This paper is focus on assembly industries actual experiment, and using the mixed integer programming(MIP) method to build our mathematical model. This model design depend on the MRP system and create the “IBM ILOG CPLEX OPTIMIZATION STUDIO” enviroment to import industries real data to experimental analysis. This model is based on the different time in production process dividing into the planned lead time, actual release time and actual production time. Specially consider the materials character and production conditions applies to fixed production time period to find the planned lead time optimal solution to help manager make appropriate decision in the real enviroment limits.
