以資料探勘方法改善製造業物料庫存呆滯之管理

Data-Mining Approach to Excess and Obsolete Materials Control in Manufacturing

鄭延邦(Yen-Pang Cheng)

指導教授 ： 劉寶鈞

元智大學/資訊學院/資訊工程學系/碩士(2008年)

摘要

目前一般製造業對於呆滯料管理所使用的方法，大多透過人工定期檢視的方法來管理，且只能針對未來的需求預測是否能夠完全將目前所擁有的庫存使用完畢來判斷物料是否形成呆滯。但實際上呆滯料產生的原因，不僅由單一的原因所造成，本論文透過資料探勘的方法可以儘可能將呆滯料形成的原因找出，再經由類神經網路的學習，將不同呆滯料成因之間的關係進行分析，找出呆滯物料的特性，藉此便可依據物料當前的狀態，預測出該物料未來可能發生呆滯的機率，進而改善呆滯料管理方式。

關鍵字

資料探勘 類神經網路 呆滯料管理

並列摘要

In manufacturing, the most popular approach to manage material excess and obsolete is via manually reviewing within certain period. And they usually focus on the relation between supply and demand to monitor if the materials are able to consume completely. In fact, the root cause of the materials excess is not only one single event but multiple reasons caused by forecast changes, lead time changes, unexpected consumption etc. In this paper, we are trying to find out all the root causes of materials excess as many as possible by the method of data-mining and using the concept of artificial neural network to learn the knowledge from these root causes. Based on the knowledge, we can predict material excess with its own attributes and provide to the users for the improvement.

並列關鍵字

Data-Mining Neural Network Excess and Obsolete Materials Control

參考文獻

[1] J.A. Harding, M. Shahbaz, Srinivas, A. Kusiak, 2006 “Data Mining in Manufacturing: A Review” ASME, Nov. 2006, Vol.128 p969-97

[2] Chi Zhou, Peter C. Nelson, Weimin Xiao, Thomas M. Tirpak, 2001 “An Intelligent Data Mining System for Drop Test Analysis of Electronic Products” IEEE Transactions on Electronics Packaging Manufacturing, Jul. 2001, Vol. 24 No. 3 p222-231

[3] Andrew Kusiak, 2001 “Rough Set Theory: A Data Mining Tool for Semiconductor Manufacturing” IEEE Transactions on Electronics Packaging Manufacturing, Jan. 2001, Vol. 24 No. 1 p44-50

[4] Keki B. Irani, Jie Cheng, Usama M. Fayyad and Zhaogang Quan, 1993 “Applying Machine Learning to Semiconductor Manufacturing” IEEE Expert, Feb. 1993, p41-47

[5] Choonjong Kwakand and Yuehwern Yih, 2004 “Data-Mining Approach to Production Control in the Computer-Integrated Testing Cell” IEEE Transactions on Robotics and Automation, Feb. 2004, Vol. 20, No. 1 p107-116