隨著科技進步，商業競爭的激烈化以及商品細緻化、複雜化，為提供更良好的服務是企業競爭與生存的重要策略，適切良好的服務為商品供應商與使用者所創造的功能與價值與日俱增。在高科技製造業中，生產設備的供應商隨著設備的複雜程度提高，所提供的設備之售後服務需求也越來越高。

本研究主要目的在運用資料科學與人工智慧之概念、方法與技術，設計「智慧服務模型」，依此模型與智慧服務分析架構，再設計「智慧服務異常處理知識挖掘方法」，並以一企業實例驗證本研究所提之模型與異常處理知識挖掘方法之有效性。
本研究所提之模型架構與挖掘方法將協助設備克服快速掌握問題根因並加以解決，持續地掌握設備狀態與趨勢、預測設備可能發的問題，而加以預防；同時，能不斷地累積知識經驗、優化服務系統，建構智慧設備服務，提昇競爭力共同創造功能價值。

With the advancement of science and technology, the intensification of commercial competition and the refinement and complexity of semiconductor products. Providing better services is an important strategy for corporate competition and survival. Appropriate and good services create functions for product suppliers and users.

The main purpose of this research is to use the concepts, methods and technologies of data science and artificial intelligence to design a "smart customer service model", and design a "smart customer service exception handling knowledge mining method" based on this model and the intelligent customer service analysis framework, and verify it with an enterprise example The model architecture and mining method proposed in this research will quickly grasp the root cause of the problem and solve it, continuously grasp the status and trend of the equipment, and predict the possible problems of the equipment. , And preventive measures. Continue to accumulate knowledge and experience, optimize the service system, improve customer service of smart devices, enhance competitiveness, and create functional value together.

[1] Philip Kotler, Marketing Professional Services, 1977
[2] James Fitzsimmons, Service management: Operations, strategy, and information technology, 2008
[3] Joseph Lampel and Henry Mintzberg, Customizing, 1996
[4] 吳偉文, 服務科學與服務創新策略, 2009
[5] Lusch, R.F. & Vargo, S.L. & Wessels, G., Toward a conceptual foundation for service science: Contributions from service-dominant logic, 2008
[6] 安部忠彦, 富士通総研(FRI)経済研究所, サービスサイエンス」とは何か，研究レポートNo.246., 2005
[7] Han and Kamber, Data mining: Concepts and techniques, 2006
[8] Berry and Linoff, Mining association rules on significant rare data using relative support, 1997
[9] 李雄飛、李軍,高等教育出版社, 數據挖掘與知識發現, 2003
[10] Thomas H., Guidelines for the diagnosis and treatment of cholangiocarcinoma: an update, 2012
[11] Eric Siegel, Predictive analytics: The power to predict who will click, buy, lie, or die, 2013
[12] TechTarget BERKELEY, Analysis of online courses to acquire data science literacy, 2017
[13] New York University, Data science and its relationship to big data and data-driven decision making, 2013
[14] Howard Gardner, The theory of multiple intelligences, 1983
[15] Cambridge Dictionary, Cambridge dictionary-smart, Cambridge dictionary.
[16] Merriam-Webster, Merriam-Webster-smart, Merriam-Webster,
[17] Word Central Student Dictionary, Dictionary-smart, Word Central Student Dictionary
[18] Oxford Dictionaries, Oxford Dictionaries-smart. Oxford Dictionaries.
[19] Wordnet 2.1, Wordnet-smart. Wordnet.
[20] R.agrwal, A. Mining Association Rules Between Sets of Items in Large Databases, 1993
[21] Quinlan, Induction of decision trees, 1986
[22] Breiman et al., A bayesian cart algorithm, 1984
[23] Kass, An Exploratory Technique for Investigating Large Quantities of Categorical Data, 1980
[24] 彭金、張盛鴻、簡禎富、楊錦情, 台電配電事故定位之研究, 2005
[25] 賴香菊, 建立智慧型售後服務系統參考架構, 1997
[26] 蔡玉慈, 建立客戶售後服務價值評估模型, 2011
[27] 曹家馨, 推估相關電聯車維修設備設施的需求量，瞭解相關維修設施設備之配置, 2008