Development of an Electric Board Maintenance Supporting System based on Association Rule Mining

碩士 === 中華大學 === 資訊工程學系碩士班 === 94 === In the last century, the rail industry has placed an important position in the transportation area, and offers a great contribution to the economics. During industrial operation, the electric board failure is an inevitable problem. It often requires experienced e...

Full description

Bibliographic Details
Main Authors: CHEN HSING-BI, 陳星壁
Other Authors: Judy C. R. Tseng
Format: Others
Language:zh-TW
Published: 2006
Online Access:http://ndltd.ncl.edu.tw/handle/43142037397577034819
id ndltd-TW-094CHPI0392061
record_format oai_dc
spelling ndltd-TW-094CHPI03920612016-06-01T04:14:44Z http://ndltd.ncl.edu.tw/handle/43142037397577034819 Development of an Electric Board Maintenance Supporting System based on Association Rule Mining 以關連法則探勘為基礎之電路板件維修輔助系統 CHEN HSING-BI 陳星壁 碩士 中華大學 資訊工程學系碩士班 94 In the last century, the rail industry has placed an important position in the transportation area, and offers a great contribution to the economics. During industrial operation, the electric board failure is an inevitable problem. It often requires experienced engineers to effectively and efficiently repair the electric boards to reduce the loss that failure made. On the other hand, some failures may cause the other failures occur in the future. To avoid the possibility of malfunctions and reduce the chances an electric board need to be repaired in the near future, it is important to find out the correlations among failures and eliminate all the causes of failures at one repair cycle. In this study, the Apriori Association Rules Mining algorithms, a famous technique in Data Mining, is applied to extract the association rules of failures from the maintenance database in the Electric Board Maintenance Management System. We use the rules to design an Electric Board Maintenance Supporting System. This system analyzes the relation between the electric board failure and repair strategy according to the rules extracted from maintenance database. It not only provides the suggested repair strategy, but also offers suggestions on the maintenance policy for potential problems. These advices help engineers to take precautions, and then lower the chances of electric board failures. At present, this system is online to serve the maintenance department of some famous enterprise in Taiwan track industry. According to the data collected from online tests during 3 months, the successful rate of repairing electric board failures is increased from 73% up to 84%. This result indicates that the system we developed helps a lot in maintaining the electric board. The benefits are reducing the loss resulted from multiple failures and increasing the profit gain of the track industry. Judy C. R. Tseng 曾秋蓉 2006 學位論文 ; thesis 88 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 中華大學 === 資訊工程學系碩士班 === 94 === In the last century, the rail industry has placed an important position in the transportation area, and offers a great contribution to the economics. During industrial operation, the electric board failure is an inevitable problem. It often requires experienced engineers to effectively and efficiently repair the electric boards to reduce the loss that failure made. On the other hand, some failures may cause the other failures occur in the future. To avoid the possibility of malfunctions and reduce the chances an electric board need to be repaired in the near future, it is important to find out the correlations among failures and eliminate all the causes of failures at one repair cycle. In this study, the Apriori Association Rules Mining algorithms, a famous technique in Data Mining, is applied to extract the association rules of failures from the maintenance database in the Electric Board Maintenance Management System. We use the rules to design an Electric Board Maintenance Supporting System. This system analyzes the relation between the electric board failure and repair strategy according to the rules extracted from maintenance database. It not only provides the suggested repair strategy, but also offers suggestions on the maintenance policy for potential problems. These advices help engineers to take precautions, and then lower the chances of electric board failures. At present, this system is online to serve the maintenance department of some famous enterprise in Taiwan track industry. According to the data collected from online tests during 3 months, the successful rate of repairing electric board failures is increased from 73% up to 84%. This result indicates that the system we developed helps a lot in maintaining the electric board. The benefits are reducing the loss resulted from multiple failures and increasing the profit gain of the track industry.
author2 Judy C. R. Tseng
author_facet Judy C. R. Tseng
CHEN HSING-BI
陳星壁
author CHEN HSING-BI
陳星壁
spellingShingle CHEN HSING-BI
陳星壁
Development of an Electric Board Maintenance Supporting System based on Association Rule Mining
author_sort CHEN HSING-BI
title Development of an Electric Board Maintenance Supporting System based on Association Rule Mining
title_short Development of an Electric Board Maintenance Supporting System based on Association Rule Mining
title_full Development of an Electric Board Maintenance Supporting System based on Association Rule Mining
title_fullStr Development of an Electric Board Maintenance Supporting System based on Association Rule Mining
title_full_unstemmed Development of an Electric Board Maintenance Supporting System based on Association Rule Mining
title_sort development of an electric board maintenance supporting system based on association rule mining
publishDate 2006
url http://ndltd.ncl.edu.tw/handle/43142037397577034819
work_keys_str_mv AT chenhsingbi developmentofanelectricboardmaintenancesupportingsystembasedonassociationrulemining
AT chénxīngbì developmentofanelectricboardmaintenancesupportingsystembasedonassociationrulemining
AT chenhsingbi yǐguānliánfǎzétànkānwèijīchǔzhīdiànlùbǎnjiànwéixiūfǔzhùxìtǒng
AT chénxīngbì yǐguānliánfǎzétànkānwèijīchǔzhīdiànlùbǎnjiànwéixiūfǔzhùxìtǒng
_version_ 1718286996354367488