本研究以彩色濾光片生產系統為對象建立一數學模式，期望在考慮最大利潤下決定最佳的品質特性檢驗政策。彩色濾光片生產屬於序列式生產系統，每工作站中具多品質特性，模式中考慮的成本包含製造成本、品質檢驗成本、內部失敗成本及外部失敗成本等四類，對瑕疵品的處理可分為重工及報廢兩種。本研究除使用窮舉法求出實例的最佳解外，另提出基因演算法求解，結果顯示基因演算法能求得最佳解並有較快的時間。本研究結果可以提供給彩色濾光片工廠在訂定品質特性檢驗政策的參考，以提昇產業之競爭力。

This study develops a mathematical model to explore the inspection allocation problem for the color filter production system in order to determine the best inspection policy of quality characteristics of each station with the optimal profit. From the viewpoint of inspection allocation problem, the color filter production system belongs to a serial production with multiple characteristics. In our profit model, four kinds of cost are involved, namely, manufacturing costs, quality inspection costs, internal failure costs and external failure cost. The treatment of non-conforming products in the production consists of rework and scrap. In addition to the complete enumeration method, this study uses a genetic algorithm to solve for a practical production system. The results show that the genetic algorithm can also obtain the optimal solution as the complete enumeration method; but it obtains the solution quicker. We believe that the results of this study can serve as a reference to solve the inspection allocation problem of color filter plants for promoting their productivity.