Improvement of virtual metrology performance by removing metrology noises in a training dataset

详细信息    查看全文

• 作者：Dongil Kim (1)
  Pilsung Kang (2)
  Seung-kyung Lee (1)
  Seokho Kang (1)
  Seungyong Doh (3)
  Sungzoon Cho (1)
  
  1. Department of Industrial Engineering ; Seoul National University ; 1 Gwanak ro ; Gwanak-gu ; Seoul ; 151-744 ; Republic of Korea
  2. IT Management Program ; International Fusion School ; Seoul National University of Science and Technology ; 232 Gongneung ro ; Nowon-gu ; Seoul ; 139-743 ; Republic of Korea
  3. SAMSUNG SDS Co. ; Ltd. ; Nongseo-dong ; Giheung-gu ; Yongin ; Gyeonggi-do ; Republic of Korea
  
• 关键词：Semiconductor manufacturing ; Virtual metrology ; Noise identification and removal ; Novelty detection
• 刊名：Pattern Analysis & Applications
• 出版年：2015
• 出版时间：February 2015
• 年：2015
• 卷：18
• 期：1
• 页码：173-189
• 全文大小：1,534 KB
• 参考文献：1. Kang P, Lee H, Cho S, Kim D, Park J, Park C, Doh S (2009) A virtual metrology system for semiconductor manufacturing. Expert Syst Appl 36:12554鈥?2561 CrossRef
  2. Kourti T, MacGregor JF (1995) Process analysis, monitoring and diagnosis, using multivatiate projection methods. Chemom Intell Lab Syst 28:3鈥?1 CrossRef
  3. Qin SJ (2003) Statistical process monitoring: basics and beyond. J Chemom 17:480鈥?02 CrossRef
  4. Qin SJ, Cherry G, Good R, Wang J, Harrison CA (2006) Semiconductor manufacturing process control and monitoring: a fab-wide framework. J Process Control 16:179鈥?91 CrossRef
  5. Su AJ, Jeng JC, Huang HP, Yu CC, Hung SY, Chao CK (2007) Control relevant issues in semiconductor manufacturing: overview with some new results. Control Eng Pract 15:1268鈥?279 CrossRef
  6. Cheng J, Cheng FT (2005) Application developmetn to virtual metrology in semiconductor industry. The 32nd annual conference of IEEE Industrial Electronics Society, USA, pp 124鈥?29
  7. Chang YJ, Kang Y, Hsu CL, Chang CT, Chan TY (2006) Virtual metrology technique for semiconductor manufacturing. International joint conference on neural networks, Vancouver, pp 5289鈥?293
  8. Chen P, Wu S, Lin J, Ko F, Lo H, Wang J, Yu W, Liang MS (2005) Virtual metrology: a solution for wafer to wafer advanced process control. IEEE international symposium on semiconductor manufacturing, USA, pp 155鈥?57
  9. Besnard J, Toprac A (2006) Wafer-to-wafer virtual metrology applied to run-to-run control. In: Proceedings of the 3rd ISMI symposium on manufacturing effectiveness, USA
  10. Kang P, Kim D, Lee H, Doh S, Cho S (2011) Virtual metrology for run-to-run control in semiconductor manufacturing. Expert Syst Appl 38(3):2508鈥?522 CrossRef
  11. Kim D, Kang P, Cho S, Lee H, Doh S (2012) Machine learning-based novelty detection for faulty wafer detection in semiconductor manufacturing. Expert Syst Appl 39(4):4075鈥?083 CrossRef
  12. Su YC, Lin TH, Cheng FT, Wu WM (2008) Accuracy and real-time considerations for implementing various virtual metrology algorithms. IEEE Trans Semicond Manuf 21(3):426鈥?34 CrossRef
  13. Lin TH, Cheng FT, Wu WM, Kao CA, Ye AJ, Chang FC (2009) NN-based key-variable selection method for enhancing virtual metrology accuracy. IEEE Trans Semicond Manuf 22(1):204鈥?11 CrossRef
  14. Chandola V, Banerjee A, Kumar V (2007) Outlier detection: a survey. Technical Report of University of Minnesota, USA
  15. Rousseeuw PJ, Leroy AM (1987) Robust regression and outlier detection. Wiley, New York CrossRef
  16. Zeng D, Spanos CJ (2009) Virtual metrology modeling for plasma etch operations. IEEE Trans Semicond Manuf 22(4):419鈥?31 CrossRef
  17. He Z, Deng S, Xu X (2002) Outlier detection integrating semantic knowledge. In: Proceedings of the third international conference on advances in web-age information management, Springer, London, pp 126鈥?31
  18. Mitchell M (1996) An introduction to genetic algorithms, MIT Press, Cambridge
  19. Yang J, Honavar V (1998) Feature Subset Selection Using a Genetic Algorithm. IEEE Intell Syst 13(2):44鈥?9 CrossRef
  20. Tax DMJ (2001) One-class classification. Ph.D. Dissertation, Delft University of Technology, Netherland
  21. Barnett V, Lewis T (1994) Outliers in statistical data. Wiley, New York
  22. Bishop CM (1995) Neural networks for pattern recognition. Oxford University Press, New York
  23. Jolliffe IT (1986) Principal component analysis. Springer, New York CrossRef
  24. Hastie T, Tibshirani R, Friedman J (2002) The element of statistical learning: data mining, inference, and prediction. Springer, New York
  25. Kohonen T (1995) Self-organizing maps. Springer series in information science, vol 30. Springer, Berlin
  26. Tax DMJ, Duin RPW (1999) Support vector domain description. Pattern Recognit Lett 20(11鈥?3):1191鈥?199 CrossRef
  27. Johnson RA, Wichern DW (1998) Applied multivariate statistical analysis. Prentice Hall, Englewood Cliffs, New Jersey
  28. Sch枚lkopf B, Smola AJ (1998) A tutorial on support vector regression. NeuroCOLT2 technical report NC2-TR-1998-030
  29. Kim D, Cho S (2012) Pattern selection for support vector regression-based response modeling. Expert Syst Appl 39(10):8975鈥?985 CrossRef
  
• 刊物类别：Computer Science
• 刊物主题：Pattern Recognition
  
• 出版者：Springer London
• ISSN：1433-755X

文摘

Virtual metrology (VM) has been applied to semiconductor manufacturing processes for the quality management of wafers. However, noises included in training datasets degrade the performance of VM, which is a key obstacle to the application of VM in real-world semiconductor manufacturing processes. In this paper, we develop a VM dataset construction method by identifying and removing noises. We define noises by considering both input and output variables and classify noises into fault detection and classification (FDC) noises and metrology noises, which have abnormal FDC variables and normal metrology variables, and normal FDC variables and abnormal metrology variables, respectively. We propose the construction of a VM training dataset including FDC noises and excluding metrology noises. By employing novelty detection methods, the normal/abnormal regions of FDC variables are identified. In experiments conducted on a real-world photolithography (photo) data, VM models trained with the dataset constructed by the proposed method showed the best accuracy and the most robustness.