A Key Parameters Analysis Method of the Quality Control in the Semiconductor Multiple Manufacturing Processes Based on Functional Data Analysis Method

Abstract:

The quality of the semiconductor products is defined by a series of the key performance parameters which have some certain relations to the electronic test parameters generated among the multiple manufacturing processes. Aimed at the quality control problem of the multiple manufacturing processes, a FDA (functional data analysis) method has been used and got the mapping relationship between the process parameters of the product lines and the product quality characteristic. A simple Change-Point hypothesis has been tested to analyze the data curves generated by the FDA method, and the key process variables have been found. Then, the equalization between the new test result and the old one has been verified by the Kolmogorov-Smirnov 2-sample test method. Some multiple manufacturing processes test data, which was collected from a semiconductor product workshop, has been modeled and analyzed. And the analysis results can illustrate the key factors of the process quality control in the multiple manufacturing processes and approach the reduction of the test times and the improvement of the efficiency and effectiveness of the equipment.

• Periodical: Advanced Materials Research (Volumes 139-141)
• Pages: 1660-1665
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.139-141.1660
• Online since: October 2010
• Authors: Yang Liu, Zhi Sheng Zhang, Jin Fei Shi
• Keywords: Functional Data Analysis, Multiple Manufacturing Processes, Product Quality Characteristic, Quality Control, Test Parameters

References

[1] Z.T. Wang, F. Qiao, and Q. D. Wu: Proc. of the 2006 IEEE International Conference on Automation Science and Engineering(Shanghai, China, October 7-10, 2006). Vol. 1, pp.253-269.

[2] A.J. Su, J.C. Jeng, H.P. Huang, C.C. Yu, S.Y. Hung and C.K. Chao: Control Engineering Practice, Vol. 15 (2007) No. 10, pp.1268-1279.

[3] Z.Q. Mao, W. Kang, F. Wang and P. Raulefs: Journal Process Control, Vol. 18 (2008) No. 10, pp.954-960.

[4] S.Q. Wang, X.P. Zhang and L. Chen: Journal of Zhejiang University(Engineering Science), Vol. 42(2008), No. 8: pp.1393-1398. (In Chinese).

[5] J. Wang, Q. P. He and T. F. Edgar: Journal Process Control, Vol. 19 (2009) No. 3, pp.443-456.

[6] A. A. Khan, J.R. Moyne and D.M. Tilbury: Journal Process Control, Vol. 18 (2008) No. 10, pp.961-974.

[7] S.J. Wang, Y.M. Chen, X.D. Wang, Z.Z. Li and L.C. Shi: Chinese Journal of Electronics, Vol. 9 (2000) No. 1, pp.1-5.

[8] H. Yue, T.X. Zhao, P.J. Ma, and D.F. Zhou: Acta Electronica Sinica, Vol. 27 (1999) No. 8, pp.19-22. (In Chinese).

[9] O. Ramsayj, C. J. Dalzell: Journal of the Royal Statistical Society, Series B(Methodological), Vol. 53 (1991) No. 3, pp.539-572.

[10] N. Jin, S.Y. Zhou, T.S. Chang, and H. H. Huang: Automation Science and Engineering, Vol. 5 (2008) No. 3, pp.557-562.

[11] O. Ramsayj, B.W. Silverman: Functional Data Analysis (Springer , New York, U.S. A 2006), pp.84-85.

[12] W. G. Manteigaa and P. Vieu: Computational Statistics & Data Analysis, Vol. 51 (2007) No. 10, pp.4788-4792.

[13] X.R. Chen: Science in China, Ser. A, Vol. 31 (1988) No. 6, pp.654-667.

[14] J.L. Wang and M.I. Bhatti: Chinese Journal of Applied Probability and Statistics, Vol. 14 (1998) No. 2, pp.113-121.