研究生: |
陳奎靜 Chen, Kuei-Ching |
---|---|
論文名稱: |
以連續性實驗法建立新產品的貝氏可靠度成長模型 Deriving reliability models using sequential experiment design and Bayesian theorem for new products |
指導教授: |
利德江
Li, De-Jiang |
學位類別: |
碩士 Master |
系所名稱: |
管理學院 - 工業與資訊管理學系 Department of Industrial and Information Management |
論文出版年: | 2007 |
畢業學年度: | 95 |
語文別: | 中文 |
論文頁數: | 54 |
中文關鍵詞: | 貝氏分析 、可靠度試驗 、可靠度 、可靠度成長 |
外文關鍵詞: | Reliability test, Bayesian analysis, Reliability growth, Reliability |
相關次數: | 點閱:122 下載:2 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
隨著資訊科技的日新月異與工業技術的蓬勃發展,追求高品質、低成本的產品已經成為世界潮流所趨,其中可靠度是衡量產品品質特性之一,所謂的可靠度是指正常的使用環境下,產品在特定時間內發揮預期功能的機率。正式的可靠度試驗作業包括設計、發展、鑑定與生產四大階段,其中以發展階段的試驗對可靠度之提昇尤為重要,因為此時可以盡早發掘設計弱點以求改正,促使產品可靠度成長,然而新產品在此階段的資料過少,常常無法精確地評估其可靠度,若是在產品大量製造時進行抽樣式檢驗,將不良率當作是判斷產品的好壞,往往會造成過高的檢驗成本與不良品,因此評估新產品可靠度成為本研究的主要方向。
本研究欲提出一個以貝氏理論為基礎的方法,利用試驗所得的實際資料將不確定因素數量化,並加以改良以適合少量樣本的應用,最後在產品設計不斷改良下推導出可靠度成長模型,而本研究的失效模式與可靠度結果,皆符合以大樣本測試的趨勢,即失效率下降、可靠度增加,並符合浴缸曲線的早夭期。
Because of the rising and blooming development of information technology and industrial techniques, high quality and low cost of the products have already become a worldwide trend. Reliability is a principle to measure the quality of products. Reliability is defined to be the successful probability that a component or system will perform an expectable function for a particular time in the normal environment. A formal reliability test includes four stages - design, development, qualification, and production. The development stage is the most important because companies can correct the problems of products as soon as possible and increase the reliability of products rapidly. However, the data of new products are usually too few to estimate the reliability. If companies do the sampling inspection when the new products are manufactured and regard the failure rate as a principle for the quality of products, they usually cause high inspection cost and inferior goods. Therefore estimating the reliability of new products is the main aspect of this research.
This research addresses a method which is based on Bayesian theorem and uses a small sample to numerate uncertain factors. Finally, the method can obtain a reliability growth model, decreasing failure rate, and increasing reliability after improving the products’ design continuously.
中文部份
王宗華,可靠度試驗與評估,台北市:中華民國品質管制學會,1989。
柯煇耀,可靠度保證:工程與管理技術之應用,台北市:中華民國品質管制學會,1997。
陳耀茂,可靠度數據的蒐集與整理方法,台中市:國彰出版社,1992。
陳耀茂,可靠度工程:方法與應用,台北市:五南出版社,2001。
褚崑成譯,Calabro, S. R.著,可靠性理論與實務,台北市:中興管理顧問公司,1989。
英文部份
Akman, O. and Huwang, L., “Bayes Computation for Reliability Estimation,” Transactions on Reliability, Vol. 46, No. 1, pp. 52-55, 1997.
Ebeling, C. E., An Introduction to Reliability and Maintainability Engineering, New York: McGraw-Hill, 1997.
Fulton, J. W. and Abernethy, R. B., “Likelihood adjustment: a simple method for better forecasting from small samples,” Proceedings Annual Reliability and Maintainability Symposium, pp. 150-154, 2000.
Humberto, G. P., Victor, A. T. and Christen, J. A., “A practical method for obtaining prior distributions in reliability,” Transactions on Reliability, Vol. 54, No. 2, pp. 262-269, 2005.
Kaplan, S., Cunha, G. D. M., Dykes, A. A. and Shaver, D., “A Bayesian methodology for assessing reliability during product development,” Proceedings Annual Reliability and Maintainability Symposium, pp205-209, 1990.
Robinson, D. and Dietrich, D., “A nonparametric Bayes reliability growth model,” Transactions on Reliability, Vol. 38, No. 5, pp. 591-598, 1989.
Xiong, J., Shenoi, R. A., and Gao, Z., “Small sample theory for reliability design,” Journal of Strain Analysis, Vol. 37, No. 1, pp. 87-92, 2001.