Screening of IC Silicon Wafer Manufacturing Parameter by Conducting MTS

• Main Authors: FENG,CHIEN-JUI, 馮芊瑞
• Other Authors: Hsiao-Lin Teng
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/35826051201784355461

碩士 === 中華大學 === 企業管理學系碩士班 === 102 === The lifecycle of electronic product becomes shorter in recent years, therefore, it is necessary to seek for a faster proactive test for semiconductor material to shorten the test lifecycle in order to achieve comprehensive production. As a consequence, it creates a product with a price advantage with high yield and reduce costs. In addition to actively innovate technologies and develop superior quality of material, it is indispensable to master the product quality. With the rising demand of customer s for high quality, the current manufacturing process or processing are using large amount of parameters to cross exam with the product as basis for quality determination. However, when there are increasing number of parameters which are to monitor variation, the efficiency of quality control system which storage data, analysis statistic data and determinate the result, begins slowing down. This leads to increasing cost of quality parameter analysis and testing. Testing manufacturing engineering plays a very important role in the silicon wafer manufacturing process. It is not only a huge investment on the initial production equipment, but also the number and cost of following up testers occupy relatively huge amount of human resources in the entire production. This study aim to screen the parameters by conducting MTS on the purpose to minimize the silicon wafer product testing items. Furthermore, it undergoes an experiment to confirm the efficiency of the determinate important characteristic variable is efficiency. As a result, it can provide a competitive testing process. The case study is base on the variation testing database of a company after wafer grinding, dicing and cleaning of 300mm(12 inches) single crystal silicon. The result shows that (1)MTSis able to efficiently distinguish main manufacturing parameter and secondary manufacturing parameter. (2)Taguchi experiment design can efficiently reduce number of experiment of multiple variable.