由於TFT LCD廣泛被運用於許多相關產品當中，因而此產業的需求業快速的提升。為了滿足需求，多數的產業藉由在不同地方擴增廠區及增加投入此系統數量，以提升其產能及生產力。基於此，公司需要使生產及配銷系統最佳化。TFT LCD製造供應鏈普遍可分為三個階層:前段array、中段cell及後段模組製程。TFT LCD產業與其他製造產業不同的地方在於生產鏈當中的產能。在array及cell當中，每單位的產決定於玻璃基板的數量，而在模組製程當中，每單位的產能決定於面板的數量。TFT LCD製造為一個多階層及多網點供應鏈，其需要高計算系統為其array、cell及模組程序做最佳化的運算。本研究TFT LCD產業中的多階層及多網點供應鏈，發展一套生產及配銷模型，並以基因演算法進行求解。產能使用率及利潤的最大化為此模型的主要績效指標。實驗數據指出本研究演算法優於基準方法。本研究藉實質的數據實驗及個案研究，成功驗證所提出演算法之堅實性。在此篇演算法之時間效率方面，本研究設計出一平行基因演算法進行運算。

Due to the TFT LCD manifestation into many related products, this industry has been growing rapidly along with demand increasing. To satisfy the demand, most companies have increased their production capacity and capability by increasing their number of factories in different places and causing complexity in this industry. Therefore, companies are required to optimize their production and distribution system. Thin film transistor–liquid crystal display (TFT LCD) panel manufacturing supply chain can be generally divided into three stages: array, cell, and module processes. Special characteristic of the TFT LCD manufacturing compared with other manufacturing is different capacity unit between its production chains. The capacity unit of array and cell is defined by the number of glass substrate, while the capacity unit of module is represented by the number of panel. TFT LCD manufacturing which is a multi-stage and multi-site supply chain requires high computational for optimizing the production and distribution of their array, cell, and module. This study develops a production and distribution planning model for the multi-stage and multi-site supply chain in TFT LCD industry. Particle swarm optimization and genetic algorithm are proposed in this research to solve the problem. Maximizing capacity utilization and total profit in the supply chain are become the major performance indicator in this model. Experimental results show that one of the algorithms is performs better than the other. For time efficiency of the proposed algorithm, parallel computation is designed in this research.

Chang, Y. J. (2007) “Single-stage and multi-site capacity planning under demand uncertainty for TFT-LCD industry,” M.S. thesis, Dept. Ind. Eng. and Mana.,National Tsing Hua Univ., Hsinchu, Taiwan.
Chen, C. C. (2009) “Capacity and transportation planning for a TFT-LCD supply chain,” M.S. thesis, Dept. Industrial Management, National Taiwan University of Science and Technology, Taipei, Taiwan.
Chen, T. L., Chen, Y. Y., & Lu, H. C. (2013). A capacity allocation and expansion model fit TFT LCD multi-site manufacturing. J Intell Manuf, 24, 847-872.
Chen, T. L., Lin, J. T., & Wu, C. H. (2014). Coordinated capacity planning in two-stage thin-film-transistor liquid-crystal-display (TFT-LCD) production networks. Omega, 42, 141-156.
Hassan, R., Cohanim, B., de Weck, O., and Venter, G. (2005), A comparison of particle swarm optimization and the genetic algorithm. Proceedings of the 1st AIAA Multidisciplinary Design Optimization Specialist Conference.
Kachitvichyanukul, Voratas. (2012). Comparison of Three Evolutionary Algorithms: GA, PSO, and DE. Industrial Engineering & Management Systems, 11, 215-223.
Kennedy, J., & Eberhart, R. C. (1995). Particle swarm optimization. Proceedings of the IEEE International Conference on Neural Networks, 4, 1942–1948.
Lin, J. T. & Chen, Y. Y. (2007). A multi-site supply network planning problem considering variable time buckets – A TFT LCD industry case. Int J Adv Manuf, 33, 1031-1044.
Pacheco, P. S. (2011). An introduction to parallel programming. Elsevier, Burlington.

Quinn, M. J. (2003). Parallel Programming in C with MPI and OpenMP. McGraw–Hill, New York.
Rocha, I. B. C. M., Parente Jr., E., & Melo, A. M. C. (2014). A hybrid shared distributed memory parallel genetic algorithm for optimization of laminate composites. Composites Structures, 107, 288-297.
Singer, P. (1994). Flat panel displays: an interesting test case for the US. Semiconductor International, 17, 78–88.
Shankar, B. L., Basavarajappa, S., Chen, J. C. H., Kadadevaramath, R. S. (2013). Location and allocation decisions for multi-echelon supply chain network – A multi objective evolutionary approach. Expert System with Applications, 40, 551-562.
Umbarkar, A. J. & Joshi, M. S. (2013). Dual population genetic algorithm (GA) versus OpenMP GA for multimodal function optimization. International Journal of Computer Applications, 19, 0975-8887.
Wang, K. J. & Lee, C. H. (2015). A revised ant algorithm for solving location-allocation problem with risky demand in a multi-echelon supply network. Applied Soft Computing, 32, 311-321.
Wang, K. J., Makond, B., & Liu, S. Y. (2011). Location and allocation decisions in a two-echelon supply chain with stochastic demand–A genetic-algorithm based solution. Expert Systems with Applications, 38, 6125-6131.
Zegordi, S. H., I. N. Abadi Kamal, and M. A. Nia Beheshti. (2010). A novel genetic algorithm for solving production and transportation scheduling in a two-stage supply chain. Comput. Ind. Eng., 58, 373-381.
Zhu, H., Pu, C., Eguchi, K., Chen, S., Liu, Yu., and Zhang, S. (2009), Paralleling genetic annealing algorithm with openMP. Second International Conference on Intelligent Networks and Intelligent Systems, 681-684.