在半導體產業中，線上製程通常包含了300個以上的製造步驟與檢測站，使得整體的生產流程時間相當長，線上各檢測站抽檢時更包含了重工、報廢等回流問題使得整體流程之抽樣策略決策比起一般製造業更具挑戰性。當提升各站抽樣頻率時，雖然可以降低不良品加工浪費所造成的成本，但相對增加了生產週期時間使產出降低進而導致成本上升，如何權衡抽樣風險與生產週期時間達最佳抽樣策略為本研究探討的主題。在本研究中，我們考慮了線上各製程與各檢測站作業時間之隨機性，發展出一個數學模型來描述此問題。基於此數學模型，本研究建立模擬模型並透過隨機最佳化演算法與多重起始解求解進行比較，透過權衡因抽樣策略所造成之不良品後續加工成本及生產週期時間增加所造成之成本，達到最小化總成本。數值實驗與實證研究充分的證明了本研究所提出之最佳化半導體線上抽樣策略於實務上的可行性。

In semiconductor manufacturing, the process is one of the most complicated manufacturing processes in the world. The process usually consists of more than 300 manufacturing steps that have complex interaction with each other. Frequent sampling can lead to redundant tests and increased cost, while infrequent sampling makes the quality of final products doubtable. There is a nontrivial tradeoff. How to decide on the optimal sampling policy is a critical, but challenging problem, in semiconductor manufacturing. In this thesis, we study the sampling problem in semiconductor industry and develop a mathematical model to characterize it. We use the simulation optimization technique to solve this model, and compare with multiple start schemes. We also derive the optimal sampling policy that can achieve minimum cost. A numerical experiment and an empirical study are conducted to verify the viability of the proposed model.

摘要 I
ABSTRACT II
目 錄 IV
圖目錄 VI
表目錄 VII
一、 緒論 1
1.1 研究背景與動機 1
1.2 研究目的 2
1.3 研究流程及論文架構 3
二、 文獻探討 5
2.1 抽樣計畫 5
2.2 半導體抽樣計畫 8
2.3 模擬最佳化 9
三、 問題定義 13
3.1 問題描述 13
3.2 數學模型 14
四、 研究方法 16
4.1 COMPASS演算法 17
4.2 多重起始解 20
五、 數值實驗 22
六、 實證研究 28
七、 結論與未來研究 34
7.1 結論 34
7.2 未來研究 35
參考文獻 36

[1]. Chang, K.H., Hong, L.J., and Wan, H. 2013. Stochastic trust-region response-surface method (STRONG)-a new response-surface framework for simulation optimization. INFORMS Journal on Computing, 25(2), 230-243.
[2]. Chien, C.F., Hsu, S.C., Peng, S., and Wu, C.H. 2000. A cost-based heuristic for statistically determining sampling frequency in a wafer fab. In Semiconductor Manufacturing Technology Workshop, 2000. 217-229.
[3]. Chien, C.F., and Hsu, C.Y. 2011. UNISON analysis to model and reduce step-and-scan overlay errors for semiconductor manufacturing. Journal of Intelligent Manufacturing, 22(3), 399-412.
[4]. Fu, M.C. 2002. Optimization for simulation: Theory vs. practice. INFORMS Journal on Computing, 14(3), 192-215.
[5]. Glover, F., and Laguna, M. 1999. Tabu search . Springer US.
[6]. Goldberg, D.E. 1989. Genetic algorithms in search, optimization, and machine learning. Reading Menlo Park: Addison-wesley.
[7]. Good, R.P., and Purdy, M.A. 2007. An MILP approach to wafer sampling and selection. IEEE Transactions on Semiconductor Manufacturing, 20(4), 400-407.
[8]. Hong, L.J., and Nelson, B.L. 2006. Discrete optimization via simulation using COMPASS. Operations Research, 54(1), 115-129.
[9]. Kirkpatrick, S., and Vecchi, M.P. (1983). Optimization by simulated annealing. Science, 220(4598), 671-680.
[10]. Myers, R.H., and Anderson-Cook, C.M. 2009. Response surface methodology: process and product optimization using designed experiments. John Wiley & Sons.
[11]. Nurani, R.K., Akella, R., and Strojwas, A.J. 1996. In-line defect sampling methodology in yield management: an integrated framework. IEEE Transactions on Semiconductor Manufacturing, 9(4), 506-517.
[12]. Nurani, R.K., Strojwas, A.J., Maly, W.P., Ouyang, C., Shindo, W., Akella, R., and Derrett, J. 1998. In-line yield prediction methodologies using patterned wafer inspection information. IEEE Transactions on Semiconductor Manufacturing, 11(1), 40-47.
[13]. Shi, L. 2000. Nested partitions method for stochastic optimization. Methodology and Computing in Applied Probability, 2(3), 271-291.
[14]. Spall, J. C. 2005. Introduction to stochastic search and optimization: estimation, simulation, and control. John Wiley & Sons.
[15]. Tekin, E., and Sabuncuoglu, I. 2004. Simulation optimization: A comprehensive review on theory and applications. IIE Transactions, 36(11), 1067-1081.
[16]. Tirkel, I., and Rabinowitz, G. 2012. The relationship between yield and flow time in a production system under inspection. International Journal of Production Research, 50(14), 3686-3697.
[17]. Tirkel, I., Reshef, N., and Rabinowitz, G. 2009. In-line inspection impact on Cycle Time and Yield. IEEE Transactions on Semiconductor Manufacturing, 22(4), 491-498.
[18]. Wan, H., Ankenman, B.E., and Nelson, B.L. 2006. Controlled sequential bifurcation: A new factor-screening method for discrete-event simulation. Operations Research, 54(4), 743-755.
[19]. Xu, J., Nelson, B.L., and Hong, J.E.F.F. 2010. Industrial strength COMPASS: A comprehensive algorithm and software for optimization via simulation. ACM Transactions on Modeling and Computer Simulation (TOMACS), 20(1), 3.
[20]. Yan, D., and Mukai, H. 1992. Stochastic discrete optimization. SIAM Journal on Control and Optimization, 30(3), 594-612.
[21]. 林則孟, (2001) , “系統模擬理論與應用”, 初版, 滄海書局。
[22]. 姜林杰祐, 張逸輝, 陳家明, 黃家祚, (2001), “系統模擬eM-Plant(SiMPLE++)操作與實務”, 初版, 華泰文化事業股份有限公司。
[23]. 鄭春生, (2010), “品質管理：現代化觀念與實務應用”, 第四版, 全華圖書股份有限公司。