APPLICATION OF PARTICLE SWARM OPTIMIZATION-BASED CLUSTERING METHOD TO REDUCE SMT SETUP TIME FOR INDUSTRIAL PC MANUFACTURER IN TAIWAN

作者：Kuo Ren Jieh^*; Lin Fang Jun

来源：International Journal of Innovative Computing Information and Control, 2010, 6(8): 3381-3400.

• Clustering analysis
• SMT production system
• ART2 neural network
• Particle swarm optimization algorithm

This study intends to propose a two-stage clustering algorithm which consists of adaptive resonance theory 2 (ART2) neural network and binary particle swarm K-means optimization (BPSKO) algorithm for grouping the orders together in order to reduce the SMT setup time. The BPSKO algorithm integrates both the particle swarm. optimization algorithm and K-means algorithm. Besides, roulette selection operator is applied for avoiding premature convergence. Simulation results using four data sets, Iris, Wine, Vowel, and Glass are very promising. The results for an international industrial personal computer (PG) manufacturer show that the proposed algorithm, ART2+BPSKO, is superior to continuous particle swarm optimization algorithm. Through order clustering, scheduling orders belonging to the same cluster together can, reduce the production time as well as the machine idle time.

• 出版日期2010-8