Forecasting production performance using data mining technique in the automobile part industry

• Main Authors: Chao-Jen Shih, 施釗仁
• Other Authors: none
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/58787394763203491956

碩士 === 國立雲林科技大學 === 資訊管理系碩士班 === 95 === There are many production variables in the processes of manufacturing which are beyond our control, such as scheduling, equipment availability, HR supply, etc. They will bring in obvious impacts to both production performance and production quality. But, can we identify what is the relationship among variables and what are the key factors? In the research, we try to find out the relationship among the production variables, as mentioned above, which do play key roles in the performance of manufacturing processes using data mining. And then, try to out-extract the key factors from the variables studied. The study plans to conduct the relevant analysis using the approaches of artificial neural network (ANN) and Decision Tree (DT). Decision Tree approach is very outstanding in the fields of classification and forecasting. We can easily obtain the classification or forecasting reports through this tool. On the other hand, ANN possesses robust computing power, and it has been widely applied in the fields of forecasting. In verification, this study collects the in-practice process data and HR data to support our inference. From the results, the accuracy for both DT approach are over 90 % and ANN approach 80%.