臺灣博碩士論文加值系統

本篇論文提出一個方法只需要最少的實驗次數即可有系統的找出最佳化的製程參數且降低研發製程的成本。此方法利用直交表來設計實驗，以類神經網路來建構直交表實驗的製程模型且結合田口式基因演算法來搜尋製程模型的全域最佳解，以達到製程參數的最佳化。此方法稱為混合式最佳化演算法，首先，利用最少的水準數來設計初始的直交表實驗，以類神經網路對此直交表建立模型，接著增加因子的水準數重新設計直交表，挑選其中的實驗點加入重新建模以改善製程模型。重複上述的步驟直到達到終止條件為止，此時田口式基因演算法所搜尋出的結果即為全域最佳值。此混合式最佳化演算法可以有效的節省研發成本且縮短研發的時間又可達到製程最佳化的目的。
本研究選用銅化學機械研磨來驗證混合式最佳化演算法的有效性。此化學機械研磨機台的可控因子包括：研磨壓力、研磨墊轉速、晶圓轉速及研磨時間。利用所設計的直交表來訓練類神經網路以建立製程模型，再以田口式基因演算法來搜尋全域最佳解，以達成最佳化參數設計。銅膜移除量的目標值設為5500 Å，將求得的最佳化參數輸入化學機械研磨機台得到實際輸出平均值為5431 Å。在此研究中只實際操作了13次實驗，但田口方法卻執行了16次實驗，此結果顯示本研究所使用的方法在求最佳化參數的問題時，相較於傳統的田口方法有更加的準確度。

This thesis presents a systematic and cost-effective approach for process optimization with minimal experimental runs. Based on the experimental design scheme of orthogonal arrays, artificial neural network is used to establish the process model. Moreover, Taguchi-genetic algorithm (TGA) is used to search for the global optimum of the fabrication conditions. The procedure starts planning and conducting the initial experiment with fewer levels. By adding experimental points selected from augmented orthogonal arrays, the process model is corrected. This step is continued until the termination condition has been reached. Then, the optimum given by Taguchi-genetic algorithm is the final solution. This proposed approach provides an effective and economical solution for process optimization.
In this research, we chose copper CMP process for verifying the effectiveness of hybrid optimization algorithm. The controllable factors of CMP machine includes back pressure, platen speed, carrier speed, and polishing time. We used orthogonal array (OA) experiments to train a neural network (NN) for creating the process model. Then we used Taguchi-genetic algorithm to find the global optimum of the control parameters. The removal rate target of the Cu film was 5500 Å. Applying the optimal parameters to the CMP machine, we got an average removal rate of 5431 Å. The result approved that the approach in this research was able to get a set of optimal parameters with better accuracy than Taguchi method.

摘要 Ⅰ
ABSTRACT Ⅱ
誌謝 Ⅲ
Contents Ⅳ
List of Tables Ⅵ
List of Figures Ⅶ
1. Introduction 1
2. Preliminaries 5
2.1 The concept of Taguchi method 5
2.2 Neural Network 7
2.3 Genetic algorithm 11
3. Hybrid Optimization Algorithm 15
3.1 Procedure of hybrid optimization algorithm 15
3.2 Experimental design by orthogonal arrays 17
3.3 Process modeling by RBFN 20
3.4 Global optimum searching by genetic algorithm 21
3.5 Selection of new experimental point or termination 21
3.6 An Illustrative Example 23
4. Industrial Applications 37
4.1 Selection of experiment 37
4.2 The importance of CMP 38
4.3 The process of CMP 39
4.4 Equipment and process parameters of CMP 40
4.5 Experiments and results 41
4.5.1 Equipment and materials 41
4.5.2 Design of experiments and results 42
5. Conclusion 47
Reference 49



List of Figures
Figure 2-1. A neuron........................................................................................7
Figure 2-2. Radial basis function network.......................................................9
Figure 3-1. The flow chart of TGA................................................................16
Figure 3-2. Flowchart of hybrid optimization algorithm...............................24
Figure 3-3. Modified Himmelblau function...................................................25
Figure 3-4. Response surface and contour of three-level factorial design.....27
Figure 3-5. Response surface and contour of batch 1....................................28
Figure 3-6. Response surface and contour of batch 2....................................29
Figure 3-7. Response surface and contour of batch 3....................................30
Figure 3-8. Response surface and contour of batch 4....................................31
Figure 3-9. Response surface and contour of batch 5....................................32
Figure 3-10. Response surface and contour of batch 6..................................33
Figure 3-11. Response surface and contour of batch 7..................................34
Figure 3-12. Response surface and contour of batch 8..................................35
Figure 4-1. The CMP machine.......................................................................41
Figure 4-2. The convergence curve of fitness value of TGA........................45

List of Tables
Table 3-1. L8(27) Orthogonal Array...............................................................18
Table 3-2. L18(37) Orthogonal Array..............................................................19
Table 3-3. The result of Himmelblau function..............................................36
Table 4-1. The control factors of copper CMP..............................................38
Table 4-2. The result of ...........................................................................43 9L
Table 4-3. The orthogonal array of..........................................................44 16L
Table 4-4. The result of experiments.............................................................45
Table 4-5. Compare with Taguchi method....................................................46

1.M. Quirk and J. Serda, Semiconductor Manufacturing Technology, Prentice Hall, 2000.
2.G. S. Peace, Taguchi Methods: a hands-on approach, Addison-Wesley, 1993.
3.D.C. Ko, D.H. Kim, B.M. Kim, and J.C. Choi, “Methodology of perform design considering workability in metal forming by the artificial neural network and Taguchi method,” Journal of Materials Processing Technology, No.80, pp.487-492, 1998.
4.D.C. Ko, D.W. Kim, and B.M. Kim, “Application of artificial neural network and Taguchi method to perform design in metal forming considering workability,” International Journal of Machine Tools & Manufacture, No.39, pp.771-785, 1999.
5.G. Casalino, F. Curcio, and F.M.C. Minutolo, “Investigation on Ti6Al4V laser welding using statistical and Taguchi approaches,” Journal of Materials Processing Technology, No.167, pp.422-428, 2005.
6.J.F.C. Khaw, B.S. Lim, and L.E.N. Lim, “Optimal design of neural networks using the Taguchi method,” Neurocomputing, No.7, pp.225-245, 1995.
7.J. Chen, D. S. H. Wong, and S. S. Jang, “Product and Process Development Using Artificial Neural-Network Model and Information Analysis,” AIChE J., Vol. 44, No. 4, pp. 876-887, 1998.
8.J. Chen, P. P.T. Chu, D. S. H. Wang and S.S. Jang, “Optimal design using neural network and information analysis in plasma etching,” American Vacuum Society, Vol. 17, No. 1, pp.145-153, 1999.
9.S. Walczak and N. Cerpa, “Heuristic principles for the design of artificial neural network,” Information and Software Technology, Vol.41, pp.107-117, 1999
10.C. Srivaree-ratana, A. Konak, and A.E. Smith, “Estimation of all-terminal network reliability using an artificial neural network,” Computers & Operations Research, Vol. 29, pp.849-868, 2002
11.J. Z. Chu, S. S. Shieh, S. S. Jang, C. I. Chien, H. P. Wan and H. H. Ko, “Constrained optimization of combustion in a simulated coal-fired boiler using artificial neural network model and information analysis,” Elsevier Science. Ltd., Fuel. 82, pp. 693-703, 2003.
12.S.P. Niculescu, “Artificial neural networks and genetic algorithms in QSAR,” Journal of Molecular Structure (Theochem), Vol. 622, pp.71-83, 2003
13.H.C. Liao, “A data envelopment analysis method for optimizing multi-response problem with censored data in the Taguchi method,” Computer & Industrial Engineering, No. 46, pp.817-835, 2004
14.C.T. Su, J.T. Wong, and S.C. Tsou, “A Process Parameters Determination Model By Integrating Artificial Neural Network and Ant Colony Optimization,” J. Chinese Institute of Industrial Engineers, Vol. 22, No. 4, pp. 346-354, 2005.
15.M. Tatilier, H.K. Cigizoglu, and A.E. Senatalar, “Artificial neural network methods for the estimation of zeolite molar compositions that form from different reaction mixtures,” computers and Chemical Engineering, Vol.30, pp.137-146, 2005
16.A.G. Olabi, G. Casalion, K.Y. Benyounis, and M.S.J. Hashmi, “An ANN and Taguchi algorithms integrated approach to the optimization of laser welding,” Advances in Engineering Software, No. 37, pp. 643-648, 2006
17.T.H. Hou, C.H. Su, and H.Z. Chang, “Using Neural Networks and Immune Algorithms to Find the Optimal Parameters for an IC Wire Bonding Process,” Expert Systems with Application, No. 34, pp. 427-436, 2008.
18.C. T. Lin and C.S. Lee, Neural Fuzzy Systems: A Neuro-Fuzzy Synergism to Intelligent Systems, Prentice Hall, 1999.
19.D.S. Broomhead and D. Lowe, “Multivariable functional interpolation and adaptive networks,” Complex Systems, vol. 2, pp. 321-355, 1988.
20.C.R. Reeves and J.E. Rowe, Genetic Algorithms: Principles and Perspectives, Kluwer Academic Publishers, 2002.
21. Century Semiconductor Technology Planarization, Semitracks Inc. , 2001.
22.A. Verikas and M. Bacauskiene, “Using Artificial Neural Networks for Process and System Modeling,” Chemometrics Intell. Lab. Syst., Vol. 67, No. 2, pp. 187-191, 2003.