本研究乃針對包含了一個工廠、多個物流中心以及多個零售商之供應鏈,進行三階補貨暨配送規劃模式探討。在給定零售商每日的訂單資訊下,進行補貨暨配送的總體最小成本之整合性考量,以決定相關的作業層次管理決策。本研究問題之主要特性為探討横向調度(Lateral Supply)問題,考量藉著各物流中心的車輛及存貨之互相支援以降低總體運輸成本。我們首先將此問題構建為一混合整數規劃數學模式,然因其具有NP-hard特性,於實務應用上有所限制,經由CPLEX最佳化軟體實際測試後發現大部分測試題的求解時間超過10小時仍常無法達到最佳解。故進而提出一結合類似運輸問題模式與最小成本網路流量模式兩個最佳化子模式之啟發式演算法,與最佳化混合整數規劃模式之求解品質與計算時間進行比較分析。實驗測試證實橫向調度無論是針對車輛或貨品以至於兩者兼具且合作的調度方式,皆顯示出此調度模式對總體成本之降低是有助益的。
This research focuses on a problem of replenishment and distribution for a three-echelon distribution system including a single plant, multiple distribution centers, and multiple retailers. Based on daily orders from retailers, we tend to suggest related operational plans according to the criteria of minimizing inventory replenishment and routing cost. Our study focuses on how to transfer inventory and vehicles among distribution centers to save total transportation costs. A mixed integer linear programming model is formulated first. Due to practical needs, we then develop a heuristic to solve problem of real world size, which is embedded with two optimization models: a variant of transportation model and a variant of minimum cost network flow model. Several small size problems were solved optimally in order to justify the performance of our heuristic. Through testing problems of 100 customers, we testified the mechanism of laterl supply is workable and may lower total transportation costs.