虛擬量測的研究與應用已日漸成熟，一套虛擬量測系統的成功與否主要取決於預測精度與預測即時性，而預測精度的好壞則直接受到資料品質的影響。本論文提出一個適用於全自動虛擬量測系統的新穎量測資料品質評估架構，本架構包含了量測機差補償及量測資料評估架構偵測力的提升，其目的為確保量測資料的品質。
量測機差的現象普遍出現於產能龐大的高度自動化生產之生產線，量測機台日以繼夜地進行產品規格的量測以配合永不停歇的生產機台之生產，然而量測機台將隨著保養方式、使用條件、機型的差異逐漸產生量測結果的偏差，此偏差將造成製程品質的失真與預測精度的下降。本研究將以模擬Daily Monitoring的方式消彌量測機差的負面影響。量測資料評估架構偵測力的提升對於虛擬量測亦是ㄧ個重要的議題，本研究將採用MDFR(Metrology Data Filter Rule)機制與Advanced ART2分群演算法作為研究主題，輔以VMI預測誤差精度確認之概念，有效地提升異常資料的偵測率並降低誤判率。

Virtual Metrology (VM) is a method to conjecture manufacturing quality of a process tool based on data sensed from the process tool and without physical metrology operation. A mature virtual metrology system should possess promptness and accuracy of the prediction. However, the VM prediction accuracy is directly influenced by the quality of process data and metrology data.

Measurement deviations among metrology tools commonly occur in large-volume and highly-automated production lines. These measurement deviations distort metrology data quality, and this will result in poor VM accuracy. To remedy this problem, a daily monitoring method that detects and compensates the measurement deviations among metrology tools is proposed in this thesis. The Automatic Virtual Metrology (AVM) system possesses a metrology data quality evaluation scheme denoted the DQIy algorithm. This DQIy algorithm adopts the adaptive resonance theory 2 (ART2) that requires large amount of real metrology data to build the DQIy model. To release this constraint and enhance the abnormality-detection accuracy of the DQIy algorithm the MDFR (Metrology Data Filter Rule) scheme is also proposed in this thesis.

[1] 鄭芳田，全自動虛擬量測，科學發展月刊，第490期，Oct 2013。
[2] F.-T. Cheng, H.-C. Huang, and C.-A. Kao, “Dual-Phase Virtual Metrology Scheme,” IEEE Transactions on Semiconductor Manufacturing, vol. 20, no. 4, pp. 566-571, Nov. 2007.
[3] Y.-T. Huang, and F.-T. Cheng, “Automatic Data Quality Evaluation For the AVM System,” IEEE Transactions on Semiconductor Manufacturing, vol. 24, no. 3, pp. 445-454, Aug 2011.
[4] 黃暄恆「適用於全自動虛擬量測系統之智慧型參數篩選機制」，國立成功大學製造資訊與系統研究所碩士論文學位論文，2013
[5] R. Stine, and D. Foster, Statistics for Business Decision Making and Analysis, Pearson Education, Inc. 2011, Ch. 26, pp. 665-693.
[6] 方世榮編著，統計學導論，第四版，華泰文化事業出版，2001，第10章，頁417-419 。
[7] 周昭宇，林裕章，統計品質管制 [Online]，Available: http://mx.nthu.edu.tw/~ctsu/QRC/studyQ/control_chart/control_chart.htm
[8] 房克成、林清風編著，Control Charts and Process Control/管制圖與製程管制，第七版，中華民國品質學會出版，1999。
[9] Douglas C. Montgomery and George C. Runger, Applied Statistics and Probability for Engineers, 5th Edition, John Wiley & Sons, Inc., New York, 2010.
[10] A. K. Jain, K. Nandakumar and A. Ross, “Score Normalization in Multimodal Biometric Systems,” Pattern Recognition, vol. 38, no. 12, pp. 2270–2285, Dec. 2005.
[11] R. Chakraborty. (2010, Dec. 10) Adaptive Resonance Theory (ART) [Online]. Available: http://www.myreaders.info/html/soft_computing.html
[12] D. Zeng, “Statistical Methods for Enhanced Metrology in Semiconductor/Photovoltaic Manufacturing,” EECS Technical Reports, pp. 19, dec. 2012.
[13] M. Halkidi, , Y. Batistakis, and M. Vazirgiannis, “Clustering Validity Checking Methods: part II,” ACM SIGMOD Record, vol. 31 n.3, Sept, 2002.
[14] 劉鈞霆「ART2參數自動尋優機制」，國立成功大學製造資訊與系統研究所碩士論文學位論文，2011
[15] Rebecca Su，統計生活館 [Online]，Available: http://home.educities.edu.tw/rebecca0924/stat/box/