台灣在過去幾十年來，都是以產業代工為主，但在高科技產業崛起下，台灣政府便致力於高科技產業的推動，促使台灣製造產業的轉型。在高科技產業競爭日漸白日化下，如何更有效的利用供應鏈資源，以及降低供應鏈成本已經不是企業能用來提升自己競爭力的工具，而是不得不採取的方案。因此，如何運用供應鏈來提高客戶滿意度已成為供應鏈管理的新課題。本文採用的方法是將客戶期望交貨日期(Customer Request Date)列入生產規劃的考量中，在產能允許的情況下在客戶期望交貨日期進行交貨。而為了讓生產規劃更有彈性且更省時間，我們在製程中多加了虛擬分類程序來將產品以其需求分為兩類，因而得以同時規劃CRD基礎以及承諾日期為基礎的兩種生產規劃。而本文所探討的問題情境是高科技產業中常見的多階段、多廠區之供應鏈網路問題。希望透過數學模型生產出兩種最佳供應鏈產能規劃。 本研究的產業背景是以發光二極體LED為基礎。在產能的限制下以供應鏈整體成本最小化為目標，利用「IBM ILOG CPLEX Optimization Studio」來求模型最佳解，並在最後將實際資料帶入模型中進行實驗驗證，找出兩種生產規劃的最佳解以供決策者參考。 In the past few decades, Taiwan had been dominated by OEM/ODM industries, but the rise of the high-tech industries along with Taiwan government’s commitment to promote it has prompted the transformation of Taiwan's manufacturing industry. The increasing competition in the high-tech industry had shifted the main purpose of supply chain studies in this field: using the supply chain resources more efficiently and reducing the supply chain costs can no longer be the strategies which help the business become more competitive. It had become a system that a business must accomplish to survive. Therefore, how to use the supply chain to improve customer satisfaction has become a new topic of supply chain management. In this article, we will conduct a new supply chain model which will take the customer request date into production planning consideration. If the capacity is sufficient, we can consider moving forward the delivery date from commit date to customers request date. In order to make production planning more flexible and time saving, we have added a theoretical binning process which will bin the final product along with its demand into two virtual classes. This procedure allows us to produce both CRD and traditional capacity planning simultaneously. The scenario of this paper will be based on the common problem in high-tech industries: multi-stage, multi-plant supply chain network problems. By using the mathematical model, we wish to produce two optimal supply chain capacity plans. The background of this study is based on the light-emitting diode LED industry. Under the capacity constraints, our goal is to minimize the cost of supply chain as a whole. We will be using "IBM ILOG CPLEX Optimization Studio" to seek the optimal solution of the model, and at least, input the actual data into the model for experimental verification and to produce two kinds of production solution for decision makers.
