根據客戶轉換效應發展出的動態套裝組合策略

Abstract

服務商品套裝組合在現在的e-commerce被廣為應用，網際網路的興起，不僅改變了人們的生活方式，也為商業市場注入新的元素。首先，上網人口不斷增加，許多人開始透過網路獲取資訊與人溝通，線上商品越來越多樣化，客戶也不斷在增加，大量的資料需要經過處理才能獲得廠商所需要的資訊，管理者可利用資料探勘的技術，從大量的資料中獲得更有意義的資訊，進一步去分析與研究何種模式能替廠商帶來最大的利潤，不論是在商品套裝組合內容的選擇或是最佳定價都已經有許多研究投入，然而不同的時間點會造成最佳定價策略的改變，如何動態訂定最佳價格與動態的預測顧客購買的動向，使得廠商獲得最大利潤的研究均是被需要的。本研究擬對此消費市場利用關聯式演算法將經常被消費者同時購買的商品包裝在一起，因為此種包裝策略可能隨著時間改變，所以本研究加入了時間因素，使得邊際成本在不同時間點也會跟改變，定價也隨之改變，所以必須使用動態數學模組去決定商品的定價，再利用消費者對商品之間的轉換關係與消費者對某商品願付的價格，分析預測長時間內消費者消費某項商品數目的變化，進而預測長時間內某項商品的總獲利。

Product bundling is a widespread practice in current e-commerce environment. Internet spring up, it not only has changed people's life style but also inject the new element into commercial market. First of all, the population to surf the net is increasing constantly. A lot of people begin to obtain information and communicate with people through the network, on-line products are more and more diversified , the customer is also increasing constantly, and a large amount of data need to be transformed into useful information. The administrator can utilize technology of data mining, and then obtain more meaningful information from a large amount of data. Analyze and study further which kind of way can bring the largest profit to the supplier . Many studies have examined optimal bundling and pricing in a static time domain. However, the optimal bundling strategy will change with time, so how to set the optimal price and predict customer behavior dynamically such that company will create maximum profits is needed. In this study, we plans to utilize the association algorithm to bundle the products that will be often bought at the same time on this market. Because this optimal bundling will change with time, we consider the time factor so that the marginal cost will change at different time period and pricing strategy will also change. So we must use the dynamic model and customer transition effect within products and the reservation price to analyze and predict the profits of certain product further in long time.