Application of Data Mining Technique to Materials Management of Rolling Stock / 應用資料探勘技術於軌道運輸系統之維修物料管理

碩士 / 國立高雄第一科技大學 / 運籌管理研究所 / 100 / Maintenance labors, materials and working days are the major factors to impact the rolling stock maintenance activities. Apporimated 28,000 listed materials of rolling stock in the 33 Illustrated Parts Catalog (IPC) chapters, which were installed in more than 70,000 locations of Bill Of Material (BOM) Noted that each item ID is unique and can be correctly related to the installed location and materials.

The IPC number can be referred to assign to the right installation locations of BOM or equipment. In this study, Data mining technique-Basket analysis is applied to BI & GI rolling stock materials records for the cases in the year of 2011. The minimum support and minimum confidence value are 0.01% and 70%, respectively. The related association rules of explore can be applied to material planer to review the materials demand list, and avoid the shortage situation for omission demand item.

In order to speed up the inquiry of IPC number for system operation and materials installed location accuracy, we build the matrix of related maintenance materials by their own installation IPC chapters through the accossiation rules. And then we can review and investigate the despriency of IPC number between the actual materials issued and item master data of operating system. Therefore the optimum result can feedback to item master of operating system and adjustment the IPC number accordingly.

Identiferoai:union.ndltd.org:TW/100NKIT5682033
Date January 2012
CreatorsChien-Yuan Wang, 王建元
ContributorsJiun-Yan Shiau, Feng-Ming Tsai, 蕭俊彥, 蔡豐明
Source SetsNational Digital Library of Theses and Dissertations in Taiwan
Languagezh-TW
Detected LanguageEnglish
Type學位論文 ; thesis
Format91

Page generated in 0.0014 seconds