現行半導體產業IC產品的生產係由多個專業廠商分工合作所完成，使得製造商(如IC設計公司)無法有效地控管分散於各個委外廠（如晶圓代工廠及封裝測試廠）的在製品。因此，本研究提出一個以製造商為中心的先進在製品管理系統(Advanced WIP Management System, AWMS)架構，主要包含了資料前處理、產出推估與出貨配置等三大功能模組。資料前處理模組具有資料驗證、資料萃取與轉換，以及資料品質確保等功能，可將委外廠所提供的原始在製品資料轉換成正確且易懂的資訊，以供製造商進一步分析與應用。產出推估模組可讓製造商利用委外廠所提供的在製品資訊，主動推估出委外廠的生產製造資訊，使製造商能正確、及時地進行生產管制。而出貨配置模組則可計算出最佳的出貨配置。本研究除了完成先進在製品管理系統架構之開發外，還設計與實作了其中之在製品資料前處理模組，以及各式功能之圖形使用者介面。最後我們以實際面板IC製造廠中的在製品資料進行系統功能測試，測試結果驗證了本論文所提資料前處理模組之有效性。相信本先進在製品管理系統架構將可應用於建構半導體工程鏈之在製品監控應用，進而使在製品資訊能在製造商、委外廠商及顧客間有效率地傳遞，以協助製造商有效地控管分散於各個委外廠之在製品，以及滿足客戶之需求。

In current Semiconductor industry, the production of ICs is completed collaboratively by several professional vendors, such as Design House, Foundry, and Assembly and Test House. The manufacturer, such as Design House, is incapable of effectively monitor the work in process (WIP) distributed in outsourcing factories, such as Foundry and Assembly and Test House. Therefore, this research proposes a manufacturer-centric advanced WIP management system (AWMS) framework, which includes a data preprocess module (DPM), a throughput prediction module (TPM), and a shipping allocation module (SAM). The DPM contains functions of data verification, data extraction and transformation, and data quality assurance. It can transform the raw WIP data, called historical WIP data, provided by outsourcing factories into correct and comprehensible information for the manufacturer to do further analyses and applications. The TPM will allow the manufacturer to predict the throughput information of the outsourcing factories based on the historical WIP data. In turn, the manufacturer can thus monitor and control the production correctly and in time. The SAM is able to compute optimal shipping allocation plans. In addition to the development of the AWMS framework, we design and implement the WIP data preprocess module and the graphical user interfaces of various functions. Finally, we apply real WIP data of TFT-LCD driver-IC factories to conduct functional tests. The testing results have validated the effectiveness of the developed DPM. It is believed that the proposed AWMS framework can be applied in constructing the WIP monitoring applications in semiconductor engineering chain such that the associated WIP data can be efficiently transferred among the manufacturer, the outsourcing factories, and the customers. This will facilitate the manufacturers to effectively monitor the WIP in their outsourcing factories and to satisfy the needs of their customers.

[1] 林俊男 「建構現場營運管制資訊系統資料倉儲之研究」，國立台北科技大學工業工程與管理系碩士論文，2004。
[2] Gartner, “Gartner Says Worldwide Semiconductor Revenue On Pace for 27 Percent Growth in 2004,” at Fairmont San Francisco, Press Release, Inc., 2004. [Online] Availible: http://www4.gartner.com/5_about/press_releases/asset_101627_11.jsp.
[3] Jou, K., “IC Design House Survey 2004,” EE Times, Taiwan, 2004. [Online] Availible: http://www.eetasia.com/surveys/eett_icd_tw04.ppt
[4] 2003 ITRS Factory Integration - Factory Information & Control Systems (FICS) Backup Foils, International Technology Roadmap for Semiconductors (ITRS). [Online] Availible: http://public.itrs.net
[5] 范盛維 「半導體工程鏈資料交換機制之設計與實作」，國立成功大學製造工程研究所碩士論文，2006。
[6] K. Hagen and M. E. Orshansky, “Adaptive Real-Time Work-In-Progress Tracking, Prediction, and Optimiztion System for a Semiconductor Supply Chain”, eSilicon Corporation, U.S. Patent,2004, US 6,748,287 Bl.
[7] M. E. Orshansky and K. Hagen, ” Prediction Based Optimiztion of a Semiconductor Supply Chain Using an Adaptive Real-Time Work-In-Progress Tracking System” eSilicon Corporation , 2004, U.S. Patent, US 7,218,980 Bl.
[8] 尹相志「Microsoft SQL Server 2005 設計商業智慧解決方案」，悅知出版， 2008。
[9] D.C. Montgomery, Introduction to Statistical Quality Control, U.S.A., Willy, 2001.
[10] RosettaNet, ”Specification of Cluster 7: Manufacturing Segment B:
Manage Manufacturing WO & WIP”, pp.11, 2002.
[11] RosettaNet, [Online] Availible: http://www.rosettanet.org/cms/sites/RosettaNet/
[12] M. Birbeck, J. Ducket, and O. Gauti, “Profession XML 2nd Deition,” in Wrox Press, 2001.
[13] 王宗立、張永政、洪敏雄、鄭芳田,”實做一個工程鏈管理系統架構- 以半導體產業為例”，自動化科技學會會刊，第50-60頁，民國95年12月。
[14] D. Decker, Outlook on the global Fabless Semiconductor Industry, FSA presentation, 2004.
[15] J. Chang and F.-T. Cheng, “Engineering-Chain Requirements for Semiconductor Industry,” in Proc. 2005 IEEE International Conference on Automation Science and Engineering, Edmonton, Canada, pp. 381-386, August 2005.
[16] 陳文耀 「從半導體設備工程系統到工程鏈管理系統的建置與能力模型之研究」，國立台灣大學資訊管理系碩士論文，2004。
[17] Microsoft Web site. [Online] Availible: http://www.microsoft.com/
[18] SEMI, “What Are the Challenges and Opportunities in Semiconductor IP? ”,FSA, 2004. [online] Availible: http://fsa.org/resources/presentations/SVF_III_051304.pdf
[19] R. Sturm, F. Frauenhoffer, J. Dorner, O. Kirschenhofer, and T. Reisinger, “Advanced WIP Control for Make-to-Order Wafer Fabrication,” Advanced Semiconductor Manufacturing Conference and Workshop, pp31-36, Sept. 1999.
[20] R. G. Qiu, “Virtual Production Line Based WIP Control for Semiconductor Manufacturing Systems,” International Journal of Production Economics, vol. 95, no. 2, pp. 165-178, Feb. 2005