Application of Decision Tree on Mining Factors of Product Test Fails

• Main Authors: Chao,Po-Chun, 趙柏鈞
• Other Authors: Judy C.R.Tseng
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/47924238839529984658

碩士 === 中華大學 === 資訊工程學系碩士班 === 103 === In toady's profit competitive era, the product quality and production process management is very important for High-tech Company. Because of the NG product will affect company’s cost, competitiveness, sustainable development and customer satisfaction. Customer requirements for quality are not only on products but also on environmental surfaces, operational surface, managing surface, and policy surface. So company must to implement quality certificate like as ISO27001, ISO9001 and ISO9000 to inporve all of quality surface needs. Due to the multivariate product or process complexity, the problem of product is difficult to be found out quickly and easily. Engineers unable to meet customer requirments for providing the reasons and solutions of product test fails in a short period of time although there are various methods of Quality Analysis like as 5-Why, 8D, Fishbone Diagram. Therefore, to solve many problems are required to rely on the experience of engineers. But it also brings another serious problem on keeping technical knowledge. Data Mining is an analytical method can be applied to find problems associated. For large amounts of data generated during production line process. It can be fast and efficient analysis relations for problems in correlation.It can alternative engineering personnel to analyze difficult problems and improve troubleshooting efficiency. It can also solve the problem of knowledge retention and inheritance on. Therefore, the Data Mining analysis tools and techniques can be promoted, trained, and applied within the company. In this paper, we use the company's production line data. Application of decision tree on mining factors of product test fails. Discovery the problems relation rules on environmental surfaces, work surface, production surface, and policy surface by data mining. And then, to provide the mining results of the analysis to the management units to improve decision-making for troubleshooting in shorten time. The final purpose is to enhance the company's competitiveness and customer satisfaction. Keywords: Data Mining、Decision Tree、Quality Management、Yield