晶圓量測在半導體產業生產過程中為一項確保產品品質必備的品管動作；然而為進行實際量測，就須購置量測機台與耗費一部分生產週期時間才能達成。為滿足這些需求，就須耗費進行上述動作的生產成本。因此於生產規劃時，通常希望將量測頻率減至可接受的最低需求，以降低生產成本。現階段的量測方式中，通常以固定的抽測項目與頻率來進行，且當製程特性與生產機台皆為穩定時，可藉由虛擬量測之技術來監測未抽測之產品品質。除此之外，生產機台之特性會隨著時間不同而改變，倘若在固定抽測點之外製程特性或生產機台產生變異時，將因未規劃執行量測動作而無法得知實際產品品質，且無法更新或重新訓練虛擬量測之模型，導致產品良率降低之風險提升。
　　原AVM系統之功能主要著重在“將離線且具延遲特性之品質抽檢改成線上且即時之品質全檢” 。為解決上述問題，本論文運用AVM系統中之各式指標，研發一套智慧型取樣決策機制 (Intelligent Sampling Decision Scheme, ISD Scheme)，透過ISD機制動態地判斷是否需要執行實際量測，以發揮AVM系統的優點與特性於量產環境，使其能更進一步地減少實際量測的頻率，致可減少量測機台的購置與減少生產周期時間(Cycle Time)，進而達到降低生產成本的最終目的。

　　Wafer measurement plays a significant role in monitoring the quality of wafer manufacturing. To do real metrology, it requires measuring tools and additional cycle time; hence, it is a costly and time-consuming operation. Therefore, reducing sampling rate to lowest possible is a priority for many factories in order to reduce cost and cycle time. The most common way for inspecting process quality is to apply periodic sampling inspection during manufacturing. If a manufacturing process is stable, then virtual metrology may be applied for monitoring the quality of wafers while no real metrology is available. Nevertheless, if a production variation appears during the process between periodic samplings, which no metrology measurement is provided for checking the real quality of wafers and retraining the VM model, then the risk of having less yield rate will be raised.
　　The functions of the original AVM system were emphasized on “Converting sampling inspection with metrology delay to real-time and online total inspection.” The merits of the AVM System have not been totally shown in mass-production environments, yet. One of the merits is to reduce the sampling rate for decreasing the capex and production cycle time such that the total production cost can be reduced. Therefore, this paper focuses on applying various indicators in the AVM system to develop an Intelligent Sampling Decision (ISD) scheme to reduce sampling rate.

[1]Weber, “Virtual Metrology and Your Technology Watch List: Ten Things You Should Know about This Emerging Technology,” Future Fab International, issue 22, section 4, pp. 52-54, January 2007.
[2]F.-T. Cheng, H.-C. Huang, and C.-A. Kao, “Developing an Automatic Virtual Metrology System,” IEEE Trans. on Automation Science and Engineering, vol. 9, no. 1, pp. 181-188., January 2012.
[3]J. H. Lee, “Artificial Intelligence-Based Sampling Planning System for Dynamic Manufacturing Process,” Expert System with Apps., vol. 22, pp. 117-133, Feb. 2002.
[4]J.N. Munga, S. Dauzère-Pérès, P. Vialletelle, and C. Yugma, “Dynamic Management of Controls in Semiconductor Manufacturing,” in Proc. IEEE/SEMI Advanced Semicond. Manuf. Conf., USA, pp. 1-6, May 2011.
[5]J. C.-H. Pan and D. H. Tai, “A New Strategy for Defect Inspection by the Virtual Inspection in Semiconductor Wafer Fabrication”, Journal Computers and Industrial Engineering, vol. 60, issue 1, pp. 16-24, February 2011.
[6]A. Holfeld, R. Barlović, and P. Good, “A Fab-Wide APC Sampling Application,” IEEE Trans. on Semicond. Manuf., vol. 20, no. 4, pp. 393–399, Nov. 2007.
[7]A. Boussetta and A. J. Cross, “Adaptive Sampling Methodology for Inline Defect Inspection,” in Proc. IEEE/SEMI Advanced Semicond. Manuf. Conf., 2005, pp. 25–31.
[8]R. P. Good and M. A. Purdy, “An MILP Approach to Wafer Sampling and Selection,” IEEE Trans. on Semicond. Manuf., vol. 20, no. 4, pp. 400–407, Nov. 2007.
[9]S. Dauzère-Pérès, J.-L. Rouveyrol, C. Yugma, and P. Vialletelle, “A Smart Sampling Algorithm to Minimize Risk Dynamically,” in Proc. 2010 IEEE/SEMI Advanced Semicond. Manuf. Conf., pp. 307-310, July 11-13 2010.
[10]J. Nduhura-Munga, G. Rodriguez-Verjan, S. Dauzère-Pérès, C. Yugma, P. Vialletelle, and J. Pinaton, “A Literature Review on Sampling Techniques in Semiconductor Manufacturing,” IEEE Trans. on Semicond. Manuf., vol. 26, no. 2, May 2013.
[11]D. Kurz, C.-D. Luca, and J. Pilz, “Sampling Decision System in Semiconductor Manufacturing Using Virtual Metrology,” in Proc. 2012 IEEE International Conference on Automation Science and Engineering (CASE), August 20-24 2012.
[12]D. Kurz, C.-D. Luca, and J. Pilz, “Monitoring Virtual Metrology Reliability in a Sampling Decision System,” in Proceedings of 9th IEEE International Conference on Automation Science and Engineering (CASE), March 11 2013.
[13]Y.-T. Huang and F.-T. Cheng, “Automatic Data Quality Evaluation for the AVM System,” IEEE Trans. on Semicond. Manuf., vol. 24, no. 3, pp.445-454, August 2011.
[14]F.-T. Cheng, Y.-T. Chen, Y.-C. Su, and D.-L. Zeng, “Evaluating Reliance Level of a Virtual Metrology System,” IEEE Trans. on Semicond. Manuf., vol. 21, no. 1, pp. 92-103, February 2008.
[15]S. Wold, M. Sjöström, and L. Eriksson, “PLS-Regression: A Basic Tool of Chemometrics,” Chemometrics and Intelligent Laboratory Systems, vol. 58, issue 2, pp. 109-130, Oct. 2001
[16]S. Haykin, Neural Networks, A Comprehensive Foundation, Prentice Hall, NJ:07458, 1999.
[17]F.-T. Cheng, H.-C. Huang, and C.-A. Kao, "Dual-Phase Virtual Metrology Scheme," IEEE Trans. on Semiconductor Manufacturing, vol. 20, no. 4, pp. 566-571, November 2007.