A Study of Optimal parameters for the Multi-response IC Wire Bonding Process

• Main Authors: Wei-Ding Wu, 吳唯鼎
• Other Authors: Chau-Chen Torng
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/26331066139504059210

碩士 === 國立雲林科技大學 === 工業工程與管理研究所碩士班 === 95 === For recent years, electronic products are becoming lighter, slimmer, shorter and smaller in size, and the problems about product and process are more and more complexity. Therefore, the process of experiment design of nowadays has to consider the optimization of multiple quality characteristics problem simultaneously. The high density and high pin count are required in the IC assembly process. Based on this trend, the wire bonding process is a key process in an IC chip-package industry. It is important to find the optimal parameters for the wire bonding process. Therefore, the 50μm pad pitch will become the main trend of IC assembly. Since the wire bonding process is the key process in an IC assembly, it becomes an urgent problem for IC assembly industry to improve the wire bonding capability. This research applied the genetic algorithm and response surface methodology to find the optimal parameters of multiple quality characteristics. First, the response surface methodology was used to model the regression relationship between the quality characteristics and the independent variables, and then a single measurement index was generated. Second, this single index revised overall Derringer and Suich desirability function and unconstrained desirability function, as the fitness functions of genetic algorithm. Finally, the combination of the proposed response surface methodology with the genetic algorithm was used to find the optimal parameters. This research finds that the proposed system can obtain better process parameters and responses.