運用關聯法則技術於產品開發設計之研究

Applying Association Rule Technique to Product Design

王錫中(Shi-Jung Wang)

指導教授 ： 蔡介元

元智大學/工程學院/工業工程與管理學系/碩士(2002年)

摘要

隨著資訊科技的日新月異，產品的生命週期不斷地縮短，企業唯有快速地推出新產品，才能保有競爭優勢。同步工程(Concurrent Engineering)即是針對傳統的序列式工程在開發上的缺失而衍生出的新產品開發程序管理，亦即在產品設計階段，就提前思考產品生命週期（product life cycle）中作業流程可能遇到的問題如製造、裝配、成本和可靠度等因素，進而達到縮短設計時程與減少開發成本之目的；如此一來，不但開發成本能夠降低，上市時間縮短，同時也增加了新產品的競爭力。
然而同步工程在以「客戶需求為導向」之設計整合上則著墨不多，企業若以顧客導向來發展新產品，不但可以充分瞭解顧客需求、市場競爭情形，更可以成為新產品成功的關鍵因素。基於此一原因，本研究利用關聯法則(Association Rule)的技術，分析市場中顧客之喜好產品組合資料，萃取不同的產品組合相對於客戶的影響。然而隨著新的喜好資料的不斷地產生，資料挖掘技術並無自動學習的機制，本研究將運用類神經網路(Neural Networks)，將舊的關聯規則與新的關聯規則作整合，也就是以動態的市場訊息回饋給設計人員參考，達到同步工程產品開發的運作，使得企業能對短暫的產品生命週期做一快速的反應(Quick Response)，真正掌握顧客之需求。

關鍵字

關聯法則 ； 類神經網路 ； 同步工程 ； 產品設計

並列摘要

With the help of advanced technology, product life cycle becomes shorter and shorter. Concurrent Engineering (CE), contrast to Sequential Engineering, is a product development paradigm that considers all product life cycle activities at a time to shorten design phase and lower the cost. The activities include manufacturing, assembling, reliability, and recycling. Although CE can condense time-to-market and increases competitiveness of new products, it is found that current CE practice is not enough in customer-oriented, so the design of product can’t satisfy customers’ requirements.
To solve the described problems, this research applies association rule technique to analyze the customer’s preference from different product combination of the market. Meanwhile, since the new customer purchase data occurred constantly, this research applies Neural Networks to integrate old rules with new rules. Proposed system can feedback dynamic market information to the designer so that Quick Response (QR) can be achieved.

並列關鍵字

Association Rule ； Neural Networks ； Concurrent Engineering ； Product Design

參考文獻

38. R. Agrawal and R.Srikant (1994), “Fast Algorithm for Mining Association

for Mining Association Rule in Large Databases,” Proceeding of 21th VLDB, pp. 432-444.

14. Agrawal, R. and Srikant, R. (1995), “Mining Sequential patterns,” Proceeding

of 11th Int’l Conference on Data Engineering, Taipei, Taiwan, pp. 3-14.

15. A. Savasere (1998), “Mining for Strong Negative Associations in a Large Data-