Application of data mining to improve the polarizing curl- Take T's case

• Main Authors: Tien-HsiehChen, 陳田協
• Other Authors: Der-Chiang Li
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/h5rhhg

碩士 === 國立成功大學 === 工業與資訊管理學系碩士在職專班 === 103 === This study first collects the production condition records and the quality control data on the production process of polarizers. Then, through the classification method of data mining, and analysis with the decision tree, as well as linear regression, the main factors that lead to polarizer curling are discovered and further controlled, to alleviate and control the curling problem of polarizers. Hence, the curling value will be in line with screen manufacturers’ requirements and the polarizer quality, along with the time cost of panel and polarizer manufacturers can be enhanced and saved.