假設一個生產過程中，通過篩選的產品可能被出售到主要市場或次要市場，被拒絕的產品則需要重工。在此假設下，本論文提出一個利潤模型來決定最佳的生產製程平均水準及篩選界限。當替代變數被使用來取代績效變數而導致產品在篩選時被錯誤分類的情況發生時，本文提出的方法考慮合理的懲罰成本及使用一個適合的機率模型來建立期望利潤模型以改進Lee et al. [12] 所提出的方法。本文詳細介紹尋求最佳製程平均及篩選界限的方法，並舉一例子做說明。
表單編號：ATRX-Q03-001-FM030-01 The thesis provides a new profit model to determine the optimum process mean level and screening limits for a production process, where the accepted item is sold on one of two alternate markets, and the rejected item is reworked. The proposed method uses a cost function with reasonable penalty incurred by misclassification for a product with the surrogate variable. Moreover, a suitable probability distribution is suggested to improve the performance of the method proposed by Lee et al. [12]. Methods of finding the optimum process mean level and the screening limits are given and an illustrative example is presented.
