Data Mining for Clustering Flow Time of Purchasing Process in a Small Batch Production System

• Main Authors: Su-Chen Tseng, 曾素真
• Other Authors: Chih-Sen Wu
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/53501130748564805203

碩士 === 國立成功大學 === 工業與資訊管理學系專班 === 95 === Purchasing process plays a very important role for gaining profit in manufacturing firms. Companies supply raw materials just in time in order to reduce the cost by creating a flexible, convenient, and quick purchasing process. According to different manufacturing types, processes of purchasing are varied. Make-to-order and make-to-stock are the two common practices in production systems. In a make-to-order production system, production begins after receiving orders without stock of products and raw materials, since customers’ demand are uncertain, most orders are in small batch. The small batch production has several uncontrollable factors, such as the flexible demand and crashed purchasing timing. Therefore, based on the small batch production, the thesis centers on the study of how to improve the efficiency in the purchasing process by employing the Data Mining technology. The study uses the Data Mining technology in which valid purchasing information is organized, clarified and transformed. The mining tool iDA is applied for the process of Data Mining. Data are grouped using clustering technique. The analysis is able to find out the connection between the invalid purchasing procedure, the efficiency of purchasing, and the purchase of raw materials. The findings can help companies reduce the set-up time and idle time for preparing raw materials, and hence supply raw materials in time.