透過您的圖書館登入
IP:3.14.6.194
  • 學位論文

產品零件變更計劃:應用模糊理論、田口方法及基因演算法進行零件供應商評選

Product-Part-Change Planning:Part Supplier Selection by Using Fuzzy Theory, Taguchi Method and Genetic Algorithm

指導教授 : 王河星
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


近年來市場競爭激烈、產品生命週期縮短與顧客意識抬頭,導致產品零件變更(Product Part Change;PPC)次數頻繁。當PPC計劃確定要執行時,如何在短時間內決定出最適於生產時之各零件供應商,實為現今企業最重視的問題。這些PPC所衍生的問題也是現今許多企業所要去面對的。因此,本研究首先以物料清單(Bill Of Material;BOM)進行產品的零件展開,並評估各個零件之間的組裝關係,然後透過模糊理論(Fuzzy Theory)取得每項零件供應商參數的確切值,並建構出一套符合多階產品零件的數學模式,其中考量零件供應商之成本、品質及時間,作為供應商評選參數。利用基因演算法(Genetic Algorithm;GA)進行最佳化零件供應商評選模式的求解,藉由田口方法(Taguchi method)求得演算法參數的最佳組合,並測試15組不同的權重設定之案例,以敏感度分析找出權重對於模式的影響。最後將本研究與Lingo所得結果做一比較與分析,結果顯示本研究基因演算法在求解不同例子中,部份例題所得到的解優於Lingo所求得的最佳解。本研究能提供一個完整的架構給予企業決策者在PPC之後,對於企業本身在短時間所發生零件汰換的情況下,有一個參考的依據,並提供一項快速且精確的PPC計劃。

並列摘要


In recent years, product-part-change (PPC) becomes more and more frequent thanks to the fierce market competition, shortening product life and rising consumer consciousness. When we execute PPC planning, how to use the shortest time to decide the most suitable part suppliers which becomes the serious lessons that enterprises focus on. Many exstended issues by PPC also become the major issue faced by various enterprises. Therefore, we first lists parts through Bill Of Material(BOM) in order to assess the assembling relationship of various parts. Then we get the fuzzy value of each part supplier’s parameter through Fuzzy Theory, and constructs an optimal mathematical model suitable for multi-pahse products’ parts. We are considering cost, quality level and delivery of each part supplier as selection parameters. Genetic algorithm was used to solve optimal part supplier selection model and through Taguchi method to find out the best combination of algorithm’s parameter. Furthermore, we test the 15 cases that with various weight set, and using Sensitivity Analysis to find out the weight effect on optimal mathematical model. Finally, compare the result from Genetic algorithm with Lingo and analyse them. The result shows that some of the solutions from Genetic algorithm are best than Lingo among these 15 cases. This research can provide decision makers an integrated structure after PPC, as well as being a standard when enterprises occur part change in the short time. by the way, we assisting decision-makers in acquiring the purchase information of best suppliers promptly and precisely.

參考文獻


[52]Wei, C. W., Developing a computer-aided part-change system based on graphic article, National Taipei University of Technology, Taiwan, 2005.
[1]Alan, F., "Case experience of implementing configuration management in a UK shipbuilding organization," International Journal of Project Management, Vol. 14, 1996, pp.221-230.
[2]Barbarosoglu, G., and Yazgac, T., "An application of the analytic hierarchy process to the suplier selection problem," Production and Inventory Management Journal, Vol. 38, 1997, pp.14-21.
[3]Barzizza, R., Caridi, M., and Cigolini, R., "Engineering change: A theoretical assessment and a case study," Production Planning and Control, Vol. 12, 2001, pp.717-726.
[4]Bonneville, F., Perrard, C., and Henrioud, J. M., "A genetic algorithm to generate and evaluate assembly plans," Proceedings of the IEEE Symposium on Emerging Technology and Factory Automation, Besancon, 1995, pp.231-239.

延伸閱讀