摘要

處於今日經營環境急遽變化與激烈競爭的環境中，企業深深體認到，能過瞭解顧客之需求並提供滿足顧客需求的產品或服務，以提高顧客忠誠度，企業方能保持競爭優勢；而建立顧客關係以留住顧客、創造企業利潤，則是現今企業熱切關心投入的議題。國內油品市場自2001年開放進口以來，油品業者之間競爭日益激烈，導致其獲利日益壓縮。若業者欲企圖建立永續的競爭優勢，除了積極開發新顧客外，與舊有顧客維持長久的關係，能提供滿足顧客需求的產品外，提供額外的管理服務，已日益顯得重要。
本研究以國內最大油品業者的客戶車隊其車輛燃料的消費資料為分析對象。首先，應用叢集技術以k-mean演算法將25,651個案例分成13群集。其次，以決策樹歸納分類技術之C4.5演算法做分類後並萃取分類規則或模式，其分類正確率高達97.2%以上，而每一群集之分類規則以”if……then”表示，這些規則亦即為顧客的車輛燃料消費行為或模式。所得到的這些資訊或知識能夠儲存在企業的模式庫中，供交易處理系統驗證顧客的車輛燃料消費行為是否正常，如果不符其消費模式時能及時警示並再作核對，以減少油品業者之服務人員錯誤的資料輸入或客戶異常消費行為，做好顧客的車輛管理服務，以及企業內部管理決策上例如人力資源分配的智慧應用。

Abstract

Today enterprises have realized that learning about the requirements of the customers and providing the customers with required products and services to improve the faith of the customers are two major activities to keep their advantages of competition. Establishing relationship with the customers to maintain them and create profits is the major concerns of the enterprises. Since decontrol the limitation of imports in domestic oil market in 2001, the competition between the oil operators have been increasingly intense, which increasingly shrank the profits of the enterprises. If the operators intend to establish sustainable competitive advantages, actively developing new customers, maintaining long-term relationship with the customers, as well as providing the customers with satisfactory products and extra management services to the customers, are required to serve that goal.
This study adopts the fuel consumption data of vehicle fleets of customers of the biggest domestic oil operators as the target of analysis. First, the study classifies 25,651 cases in 13 groups using k-mean algorithm of Cluster Analysis. Second, the C4.5 algorithm of decision trees induction technology is used for the classification to extract the rules or modes of classification. The tree models reach correctness rates exceeding 97.2%. The classification rule of each group is represented by ”if……then”. These rules are just the consumption behaviors or modes of vehicle fuel of the customers. The information or knowledge obtained in this way can be stored in the mode database of the enterprises for the transaction handling system to verify if the consumption behaviors of vehicle fuels of the customers are normal. Thus, in case they are not conforming to the consumption modes, the system can give alarm and check again so as to reduce the wrong inputting of data by the service personnel of the oil operators or the abnormal consumption behaviors of the customers. The system provides fine vehicle management for the customers and provides intelligent application of internal management decisions of enterprises such as allocation of human resources.

參考文獻

中文

何玉麗（2004）.自由消費˙消費自「油」──油品市場自由化與消費者權益.「能源報導」月刊,93年05月. 10.15,2005.from http://www.tier.org.tw/energymonthly/recent3.asp.
羅南.史威福(Ronald S. Swift) 著,賴士奇..等合譯,”深化顧客關係管理”,遠擎管理顧問公司,2001.
遠擎管理顧問公司編著,”顧客關係管理深度解析-執行以客戶為中心的企業轉型策略”,遠擎管理顧問公司,2001.
中國石油股份有限公司簡介. 10.11,2005. from http://www.cpc.com.tw/aboutcpc/transales.htm.

英文

Anderson,K.and Kerr,C. (2002).Customer relationship management. Mcgraw-Hill Companies,Inc.
Berry,M.J.A.and Linoff,G.(1997).Why data mining?.in Berry,M.J.A.and Linoff,G.(Ed.) Data mining techniques:for marketing,sales,and customer support.1-16.New York USA:John Wiley & Sons Inc.
Berry,M.J.A. and Linoff,G.S.（2000）.Mastering data mining,the art and science of customer relationship management. New York USA：John Wiley & Sons Inc.
Bose,I.and Mahapatra,R.K. (2001). Business data mining – a machine learning perspective. Information & Managemen, 39, 211-225.
Carbone,P.(1997). Data mining or knowledge discovery in databases: an
overview. Data management handbook, New York: Auerbach Publications.
Fayyad, U., Piatetsky-Shapiro, G., and Smyth, P. (1996). From data mining to knowledge discovery in databases. AI Magazine,17,37-55.
Fayyad, U., Piatetsky-Shapiro, G. and Smyth, P. (1996). The KDD process for extracting useful knowledge from volumes of data. Communication of ACM,39, 11,27-34.
Fayyad, U.and Stolorz,P.(1997).Data mining and KDD: promise and challenges. Future Generation Computer Systems ,13, 99-115.
Frawley,W.J., Piatetsky-Shapiro, G.and Matheus,C.J.(1991).Knowledge
discovery in databases:an overview. AAAI/MIT press.
Fisher,D.（1987）.Improving inference through conceptual clustering. AAAI Conf.,Seattle,WA,461-465.
Golfarelli, M.,and Rizzi, S., (1999). Designing the data warehouse: key step and crucial issues. Journal of Computer Science and Information Management, 2,3.
Han,J.and Kamber,M.(2000).Introduction.in Han,J.and Kamber,M.,Data mining:Concepts and techniques.1-36.USA:Morgan Kaufmann.
Han,J.(2001) Data mining:Concepts and techniques. San Diego USA:Academic Press.
Hui,S.C.and Jha,G.(2000).Data mining for customer service support. Information & Managemen, 38, 1-13.
Inmon,W.H. (2002). Building the Data Warehouse. 31-77,3thed. New York：John Willey &Sons.
Mingers, J. (1989). An empirical comparison of selection measures for decision-tree induction. Machine Learning, 3,319-342.
Michaud,P. (1997). Clustering techniques. Future Generation Computer Systems ,13, 135-147.
Quinlan,J.R.(1979), Discovering rules from large collections of examples: a case study. In Michie, D., editor, Expert Systems in the Microelectronic Age. Edinburgh Scotland:Edinburgh University Press.
Quinlan,J.R.(1993). C4.5:Programs for Machine Learning. San Mateo CA： Morgan Kaufmann.
Rajagopalan,B. and Krovi,R.（2002）.Benchmarking data mining algorithms.Database Management,13,25-35.
Shaw,J.M., Subramaniam,C., Tan,G.W. and Welge,M.E. (2001). Knowledge management and data mining for marketing. Decision Support Systems, 31,127-137.
Swift,R.S.（2001）,Accelerating customet relationships:using CRM and relationship technolongies. Prentice Hall,Inc.
The University of Waikato,WEKA soft.05.03.,2006. from http://www.cs.waikato.ac.nz/ml/weka/
Tiwana, A. (2000)The essential guide to knowledge management：e-business and CRM applications. Upper Saddle River, NJ: Prentice Hall.
Turban,E.and Aronson,J.E. (2001).Data management:warehousing,access and visualization in.decision support systems and intelligent systems. 128-164,6thed, USA:Prentice Hall.
Wayland, R.E. and Cole, P.M. (1997). Customer connections: new trategies for growth. Boston：Harvard Business school.
Witten,I.H.and Frank,E.(2000).What’s it all about?in Witten,I.H.and Frank,E.,Data mining:practical machine learning tools and techniques with java implementations.1-36.USA:Morgan Kaufmann.
Zadeh,L.A. (1965). Fuzzt sets. Information and Control,8,338-353.
Zait,M.and Messatfa,H. (1997). A comparative study of clustering methods. Future Generation Computer Systems ,13,149-159.