Tool Fault Analysis with Decision Tree Induction and Sequence Mining

Abstract:

Tool fault analysis is a common task for process engineers in modern industries to maintain high yields of the final products. Statistical process control is a monitoring method normally adopted by most engineers. Recently, there has been enormous awareness among industrial and manufacturing engineers that intelligent techniques from the data mining and machine learning fields can be applied to discover subtle patterns from the manufacturing process data. In this paper, we present the two data mining techniques, i.e. decision tree induction and sequence mining, to discover frequently occurred patterns of the low performance wafer lots in the semiconductor manufacturing industries. The comparative analysis results of both techniques are presented through experimentation over the standard data set for the purpose of re-experimentation.

• Periodical: Applied Mechanics and Materials (Volumes 548-549)
• Pages: 703-707
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.548-549.703
• Online since: April 2014
• Authors: Kittisak Kerdprasop, Nittaya Kerdprasop
• Keywords: Data Mining (DM), Decision Tree Induction, Intelligent Manufacturing, Semiconductor Process, Sequence Mining, Tool Fault Analysis

References

[1] G. May and C. Spanos: Fundamentals of Semiconductor Manufacturing and Process Control (John Wiley & Sons, 2006).

[2] S. Sarawagi, in: Advanced Methods for Knowledge Discovery from Complex Data, edited by S. Bandyopadhyay, U. Maulik, L.B. Holder, and D.J. Cook, Springer (2005), pp.153-187.

DOI: 10.1007/1-84628-284-5

[3] A. Ison, W. Li and C. Spanos: Proceedings of IEEE International Symposium on Semiconductor Manufacturing, San Francisco (1997), B-49-B-52.

[4] B. Goodlin, D. Boning, H. Sawin and B. Wise: Journal of the Electrochemical Society, 150, 12 (2003), G778-G784.

DOI: 10.1149/1.1623772

[5] G. Spitzlsperger, C. Schmidt, G. Ernst, H. Strasser and M. Speil: IEEE Transactions on Semiconductor Manufacturing, 18, 4 (2005), pp.528-533.

DOI: 10.1109/tsm.2005.858495

[6] Q. He and J. Wang: IEEE Transactions on Semiconductor Manufacturing, 20, 4 (2007), pp.345-354.

[7] G. Verdier and A. Ferreira: IEEE Transactions on Semiconductor Manufacturing, 24, 1 (2011), pp.59-68.

[8] E. Tafazzoli and M. Saif: Proceedings of American Control Conference, St. Louis (2009), pp.4329-3433.

[9] Z. Ge and H. Song: IEEE Transactions on Semiconductor Manufacturing, 23, 1 (2010), pp.99-108.

[10] M. McCann, Y. Li, L. Maguire and A. Johnston: Proceedings of JMLR Workshop, Canada (2008), pp.277-288.

[11] L. Rokach, R. Romano and Q. Maimon: Journal of Intelligent manufacturing, 19 (2008), pp.313-325.

[12] N. Ruschin-Rimini, O. Maimon and R. Romano: Journal of Intelligent manufacturing, 23 (2012), pp.481-495.

[13] Advanced Analytics – Intel: SETFI: Manufacturing data: Semiconductor tool fault isolation. Causality Workbench Repository, http: /www. causality. inf. ethz. ch/ repository. php (2008).

[14] A. Gabadinho, G. Ritschard, M. Studer and N. Muller: Mining Sequence Data in R with the TraMineR Package: A User's Guide. University of Geneva, Switzerland. (http: /mephisto. unige. ch/traminer/) (2011).