需求不確定下控檔片之管理

Abstract

半導體製作過程中控檔片(Control and Dummy Wafers) 的主要功能在於確保晶圓產品的品質與製程穩定。大量的控檔片並非產品卻不可或缺，一但短缺則會導致製程停頓與交期延誤，間接會提高成本和降低獲利。故有效的管理控檔片是晶圓製造過程中重要的議題。 本研究主要探討控檔片之降級法則與需求服務水準兩種問題，目的是在多期多產品不確定需求前提下最小化總成本。本研究以兩階段隨機規劃模型求出新片的供給數量、降級的數量、方式與途徑，探討降級法則；再以機遇限制規劃模型滿足預設的需求服務水準，並提出利用滾動時窗法將機遇限制規劃模型轉換成等價的動態線性規劃模型求解。經由實例驗證，本研究所設計之控檔片管理模式在實務上因將需求不確性列入考慮，故提高其應用上之有效性。

The first subject of this dissertation is to study a realistic planning environment in wafer fabrication for the control and dummy wafers problem (C/DWP) with uncertain demand. A two-stage stochastic programming model is developed based on scenarios and solved by a deterministic equivalent large linear programming model. The model explicitly considers the objective to minimize the total cost of C/D wafers. A real-world example is given to illustrate the practicality of a stochastic approach. The results are better in comparison with deterministic linear programming by using expectation instead of stochastic demands. The model improved the performance of C/D wafers management and the flexibility of determining the downgrading policy. For the inventory management with service level, a chanced-constrained model is developed to minimize the total cost and to keep satisfaction of customer with pre-specified probability level. Based on rolling horizon method, this model is transformed into a dynamically equivalent linear problem. A numerical example problem is illustrated to provide information for setting customer satisfaction levels and unfolding effective inventory management options.