全自動虛擬量測系統 (AVM)為一複雜之系統，其中虛擬量測伺服器 (Virtual Metrology Server, VMS)為系統運行之核心主軸。VMS具備1. 推估模型 2.資料前處理 3.監控指標等三大核心類別，各核心為能達至該功能之完整性，需事先進行大量的線下 (Off-line)實驗，才可設定適當之設定值，其過程非常冗長耗時且需投入大量專家經驗方能完成，為降低導入AVM技術之門檻，實現快速且穩定適用於產線上 (On-line)環境，本論文將提出適用於量產環境之AVM系統各核心演算法建模考量 (設定)。
自動建模為始終為現場使用者之需求目標，透過自動建模機制可降低AVM技術門檻與加速導入AVM系統於各產業生產線上的應用。然而，各核心演算法的建模設定，為自動建模機制中最困難之課題，透過本研究所提出之各核心演算法建模準則，將更能提升自動建模可行性，實現全廠自動化之虛擬量測系統應用。

Automatic Virtual Metrology (AVM) is a complex system with Virtual Metrology Server (VMS) being its core. VMS contains three modules: conjecture model, data preprocessing module, and monitoring indexes module. Each module has its own algorithms. In order to make each algorithm achieve the integrity of its functions, lots of offline experiments in advance of right setting for each core algorithm is required. The setting process of each core algorithm is very time-and-expertise-manpower consuming to complete. In order to reduce technical difficulties of AVM system implementation to achieve efficient and stable on-line environment for mass production, this paper proposes guidelines of core algorithm’s model-creation considerations (setting) for the AVM system in mass production environment.
Automatic model creation is always the objective of on-spot users. It can reduce the technical difficulties of AVM system implementation and speed up the applications of AVM on production lines of various industries. However, the core algorithm settings are the most difficult issue for automatic model creation. This paper proposes guidelines of core algorithm’s model-creation considerations (setting) to enhance the feasibility of automatic model creation and realize fab-wide VM applications.

[1] 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., Jan. 2012.
[2] 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, Nov. 2007.
[3] 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.
[4] 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.
[5] W-M Wu, F-T Cheng, and F-W Kong, “Dynamic-Moving-Window Scheme for Virtual-Metrology Model Refreshing”, IEEE Trans. on Semiconductor Manufacturing, vol. 25, no. 2, pp. 238-246, May 2012
[6] 許智誠「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」，國立成功大學製造資訊與系統研究所碩士論文學位論文，2014
[7] Pang-Ning Tan , Michael Steinbach , Vipin Kumar (2008), 資料探勘 Introduction to Data Mining , 台北, 台灣培生教育出版股份有限公司, (Orignal work published 2005)
[8] 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
[9] S. Haykin, Neural Networks, A Comprehensive Foundation, Prentice Hall, NJ: 07458, 1999.
[10] 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
[11] F.-T. Cheng, C.-F. Chen, Y.-S. Hsieh, H.-H. Huang, and C.-C. Wu “Intelligent Sampling Decision Scheme Based on the AVM System” International Journal of Production Research, published online: September 2014, DOI: 10.1080/00207543.2014.955924. http: //dx.doi.org/10.1080/00207543.2014.955924 (NSC102-2622-E-006-022-CC2, NSC102-2221-E-006-118-MY2, Project AIM-HI of MOE, ROC)