建構大量品質資料及時分析系統－以電動汽車鋰電池製造為例

Construct Real Time Mega-data Analysis System-Using Lithium Battery Production as Illustration

李世才(Shih-Cai Li)

指導教授 ： 何正斌

國立屏東科技大學/管理學院/工業管理系所/碩士(2011年)

https://doi.org/10.6346/NPUST.2011.00069

摘要

每一顆鋰電池出廠前都需經過多次且繁複的品質檢查，因而產生大量的檢驗，在分析上頗為不易。若發生品質不良的狀況，反應時間過久，造成無法預防不良之重複發生。本研究建構一套針對大量品質資料及時分析資訊系統，將機台檢驗數據直接進行資料整合，減少人為運算之錯誤及時間。此系統不僅提供品質資料之平均數及標準差，更加入了直方圖（Histogram），管制圖（Control Chart）及推移圖（Run Chart），可供使用者用者快速且明確的了解製程的變化，以降低製程分析誤判的情況並減少發生品質異常狀況之反應時間。

關鍵字

資料整合 ； 及時分析

並列摘要

Each lithium battery is subject to many complex examinations before sending to customers. It is also imperative to execute a large number of tests to assure the quality. However, the causes of defects may not be uncovered until the results of analyses are finished. Therefore, the similar defects will occur repetitively. The study is to construct real time mega-data analysis system for timely analysis of information systems that the machine test data combined with the database of the information system. The data integration can reduce human error and operation time. This system not only provides the mean and standard deviation of the quality information, but also gives the histograms, control charts and run charts that users can detect the process change very soon. The system also helps to reduce the misjudgment of the situation analysis as well as to decrease the responding time of the occurrence of abnormal quality conditions.

並列關鍵字

data integration ； real time analysis

參考文獻

1. 李柎梓，2002，「應用類神經網路於製程平均值變化之偵測」，元智大學工業工程與管理學系碩士班碩士論文。

11. Fayyad, U., Shapiro, G.P., and Smyth, P. "From Data Mining to Knowledge Discovery in Databases," AI Magazine （17：3） 1996, pp 37-54.

12. Guh, R.-S., & Shiue, Y.-R. （2008）. An effective application of decision tree learning for on-line detection of mean shifts in multivariate control charts. Computers & Industrial Engineering. 55. 475-493.

15. Montgomery, D , C., Introduction to Statistical Quality Control, 4th ed., Wilry2000.

16. Shewhart, W. A., Economic control of quality of manufactured product, Van Nostrand, New York, 1931.