[[alternative]]Application of the Association Rule of Data Mining Technology to Equipment Failure: Using Firm A as an Example

[[abstract]] 在傳統的製造業中，紡織業面對各項成本的增加，造成部分產業外移到更低廉工資成本的國家，在內部因素方面，工廠機台逐年老舊，故障次數的頻繁，設備的可靠性降低，進而影響生產效率甚鉅，造成營運成本的增加。資訊科技日新月異快速的發展，如何運用快速的方法，資料蒐集發掘潛在有用的資料，萃取成有用的資訊，形成有價值的關鍵知識。從中找出故障關聯性，並減少故障次數及提升生產效率為目標。 本論文研究探討個案公司設備故障記錄資料中，應用資料探勘技術Apriori演算法對設備故障屬性加以分析，找出顯著規則並提供改善建議，以達到降低故障發生頻率。[[abstract]] Various increasing costs in the textile industry, which is a type of traditional manufacturing industry, have resulted in the relocation of some Taiwanese factories to countries with lower wage costs. Internal factors such as yearly depreciation of aging factory machinery, frequent machine failures, and a decrease in equipment reliability have heavily impaired production efficiency and resulted in increased operating costs. Because information technology rapidly presents new developments, valuable knowledge regarding rapid data collection methods, uncovering potentially useful data, and extracting useful information has formed the field of data mining. The goals of data mining on textile production data are to find the correlations of machine failures, reduce the frequency of such failures, and enhance production efficiency. This paper is a case study that examines a firm's equipment failure records, and applies data mining technology in the form of the Apriori algorithm to analyze equipment failure attributes and identify highly significant rules. This study provides suggestions for improvement to minimize the frequency of machine failures