Simulation-Based Scheduling Design and Analysis for IC Testing Factory

• Main Authors: Hung-Cheng Chiang, 江宏正
• Other Authors: Mei-Ting Tsai
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/63694016391020489787

碩士 === 國立中興大學 === 高階經理人碩士在職專班 === 101 === According to technical advances in technology, the electronic products main appeal of today is toward slim and light, and product quality, cost and deadline are becoming the basic requirement of an industry. The competition standard is raised to speed, service and technical capacity level. Professional IC testing service plant has positive correlation with its own testing platform operating capacity and the number of testers. So Professional IC testing service plant has to purchase high-level tester to meet the expectation of the customers and the next generation product. In addition to the process capability of the tester platform, the number of testers and machine utilization level are two major profit indicators of the professional IC testing plant. Analyzing the orders the testing plant receives found that the characteristics of the orders tend to be small size, multi-model and repeated. An effective production will reduce the loss from machine change setup, and enhance competitiveness of self-production. This problem highlights the importance of scheduling and dispatching. Therefore, a better arrangement in product line can meet the customer delivery requirements, and improve customer satisfaction. Increasing output equals to the reduction in equipment investment cost and higher competitiveness. In this study, we use simulation software to do scheduling performance comparison. The simulation seeks to reduce efficiency loss from tools switching, increase output and reduce investment cost of test platform by using schedule dispatching techniques to adjust the order priority of all untested products, where it follows the logic of testing the same model device first. We use Grouping and First-In-First-Out (FIFO) rule of scheduling dispatching to run simulations. A simulation includes 6 weeks of operation time and repeats 10 times. The results showed that the Grouping has a better delivery rate than FIFO in the external performance index; also in the setup change frequency, machine utilization, the in house cycle time in system, waiting time and work-in-process queue quantity of the internal performance index. No significant difference was found in total process cycle time. The performance of this grouping scheduling dispatching is available as a reference for industry applications.