| Summary: | 碩士 === 中原大學 === 工業與系統工程研究所 === 97 === With the prosperous development in the high technology industry, the quality requirement from customers increases day by day. Meanwhile, in order to ensure product quality, the inspection of work-in-process products has become one of basic procedures. In addition, enterprises can only provide limited inspection resources in their very expensive production systems. Therefore, how to assign the limited inspection resources to the right locations is an essential issue in order to promote product quality, reduce production defect rate, and increase enterprise profit.
This thesis takes color filter production systems as the study subject to develop a mathematical model. Due to its unique multi-quality-characteristics, we need to consider not only to set up inspection stations but also to determine which characteristics to be inspected in order to obtain the inspection policy with the optimal profit. The cost considered in this model consists of process cost, inspection cost, internal failure cost and external failure cost. The defect treatment for the color filter production system has two kinds: rework after repair and scrap. The product needed to rework will return to the current workstation.
This study uses not only the complete enumeration method but also the genetic algorithm method to find the optimal solution; the latter has very similar results as the former, but uses less computation time. A sensitivity analysis of inspection policy is performed with respect to each factor of the proportion of qualified products, revenue per unit, external loss per unit, equipment depreciation and inspection equipment cost to explore the inspection policy under various conditions. We believe that the results of this study can serve as a production planning tool to solve the inspection allocation problem in color filter production systems and increase their productivity.
|
|---|
