- 學位論文
顧客關係管理應用在便利商店之研究-以校園萊爾富便利商店為例
The Study on Customer Relationship Management for Convenient Store–A Case Study of Hi-Life Convenient Store on Campus
摘要
本研究旨在探討創造商業價值中顧客關係管理帶來的契機，使企業避免花費更多成本檢定關聯法則，並執行更精確可行性產品關聯法則，以促進分析顧客消費偏好的速度，進一步獲得商業決策制定，提供創造商業價值。顧客關係管理的議題在最近幾年中快速崛起，在顧客關係管理領域中以80/20法則來看，可以得知『企業有百分之八十的利潤是由百分之二十的顧客所創造出來』，因此如何利用顧客來創造利潤對企業而言相當重要，而要找出顧客對企業的利益為何，可藉由資料探勘的技術來達成。資料探勘提供了多種的技術，其中如何得到關聯法則，是關鍵步驟，經驗告訴我們，一般很難提供真正有效用的法則，於是本研究利用隨機變數及變異數的概念，來消除關聯法則中誤導的法則及多餘的法則，以達到減少檢定法則所花費的時間，進而降低成本。 本研究以萊爾富便利商店所提供的交易資料數據為基礎，進行資料前置分析、處理，並做統計運算分析。由實證得知，在關聯法則很多時，利用本研究的方法，可以減少大約1~3成的法則，能讓便利商店能夠更精確、有效地運用所挖掘出來的法則，進而取得顧客消費偏好以進行商品組合促銷或商品陳列設計，將有效增進其便利商店的盈利收入。
並列摘要
The aim of the research is to explore the opportunities brought by the Customer Relationship Management (CRM) of creating business values. It also attempts to make enterprises avoid more costs on the survey of the Association Rule and execute the Association Rule of practicable products more precisely. Accordingly, they are able to improve the speed of the analysis of consumers’ preference, further formulate business decisions, and offer and create business values. It has been getting more and more popular about the issue of “Customer Relationship Management” for recent years. From the 80/20 Rule of CRM field, we can know that 80% of profits are made by 20% of customers for an enterprise. In other words, it is very important for an enterprise to make good use of customers to create profits. However, what is the advantage of customers to an enterprise? The answer can be found out by means of data mining. Data mining is able to provide several kinds of techniques, which how to get the Association Rule is a key to data mining. According to the earlier experiences, it would be hard to give an effective and useful rule. Therefore, this research will try to eliminate the misunderstood and unnecessary rules of Associate Rule by the concepts of random variable and it’s variation in order to lower the time and costs spent by examining rules. To carry out the research project, the author is going to do pre-data analysis, process, and statistic operation based on the transaction data from the Hi-Life Convenient Store. Due to the method of Positivism, this research can reduce 10~30% of rules in such a lot of Associate Rule. Last but not least, convenient stores can not only apply rules more accurately and effectively, but also obtain the preference of customers’ consumption to do product-combined promotion or design of product display for the increase of its revenues and profits.