透過您的圖書館登入
IP:44.223.39.199
  • 學位論文

知識管理應用在營建工程專案查核缺失分類之研究

Using Knowledge Management for Defect Classification of Construction Quality Inspection

指導教授 : 曾惠斌

摘要


每一個工程專案現場工程師都會記錄一些品質問題,在整個工程結束後,會累積許多與品質相關的Punch List (Punch List以下簡稱缺失改善單),而缺失改善單是記錄整個工程專案有關品質問題項目的表單。經過整理並電子化後,成為該工程專案的資料,這些資料往往沒有有效的分類運用,就只是保存,卻無法提高利用價值,等到需要時,發現曾經發生過的品質事件,卻不知道從哪裡發掘。要如何將資料有效分類,快速知道發生哪一類型的品質問題,就有賴知識管理的應用。然而,應用知識管理主要是建立完整的分類架構,訂分類架構卻需要大量的人力耗費,如何節省人力耗費,並依工程實際狀況建置分類項目,成為研究動機。 本研究的主要目的,將工程專案記錄上千上萬筆的缺失改善單之品質問題更自動資訊化,節省大量的人力資源,並加強工程品質相互影響的關係,協助工程師面對不同的工程專案,能自動取得延伸的關鍵詞彙與分類的品質項目。 研究成果,利用Data Mining的方式,將記錄有品質事項的缺失改善單,針對品質問題做斷詞處理,並做統計,挑選出較常出現的關鍵詞彙,常出現的關鍵字其代表可能是較常發生的品質問題,相對也較為重要。再利用詞彙相關係數計算公式,針對所作之斷詞,求得相關程度,加強缺失改善單之品質問題間的關聯性。針對缺失改善單之品質問題做斷詞及關聯計算後,擷取關鍵詞彙,並提出缺失改善單之品質問題的分類項目,如此讓品質問題有類別可循。 最後,建置分類項目,再利用實際工程專案來做驗證。本研究利用某電廠的工程專案所記錄的缺失改善單之分類項目做分類,來驗證分類項目是否符合實際工程專案狀況。再請教專家來作驗證,解釋其所建置之分類項目符合工程專案的缺失改善單之品質問題之需求。

並列摘要


Punch List is the record of construction quality inspection. The usage of Punch List, the data of engineering projects, is often not effective enough in terms of classification. To categorize the data received and identify the quality problems effectively relies on the application of knowledge management. The motivation of this study is to establish a comprehensive classification structure in accordance with the situations happened during each engineering project to reduce the human capital. The purpose of the study is to establish a methodology for retrieving information more conveniently, to save labor, and to enhance the interactive relationship of the construction quality inspection. This study uses the method of data mining for classifying the defects resulting from the construction quality inspections. Firstly, the Punch List is used to retrieve key words and to gather statistics. A formula of vocabulary correlation coefficient is used to enhance the interactive relationship of the Punch List. The classification structure of Punch List is then constructed. Finally, the proposed classification is applied in a real case to test and verify. Consultation with experts helps explain the relationship between the classification results and the Punch of construction quality inspection.

參考文獻


參考文獻
1.Salton G., McGill M. (1983),“Introduction to Modern Information Retrieval,”McGraw-Hill New York.
2.Frawley W. J., Paitetsky-Shapiro, and Matheus G.C.J. (1991),“Knowledge Discovery in Database: An Overview”, California, edited by G.Piatetsky-Shapiro and Frawley, W. J., AAAI/MIT Press, pp.1-30
3.Agrawal, R., Imielinski, T., and Swami, A. (1993),“Mining Association Rules between Sets of Items in Very Large Database,” In Proceedings of the ACM SIGMOD Conference on Management of Data, pp. 207-216.
4.Wiig, K. M. (1997),“Knowledge management: where did it come from and where will it go?”, Expert Systems With Application, Vol.13, No.1, pp.1-14

延伸閱讀