新產品的研發階段只有很少的研發產品可供使用，如何在短時間內利用有限的樣本來進行新產品的壽命預估，Tseng and Wen（2000）提出逐步應力加速衰變試驗（Step-Stress Accelerated Degradation Test，簡稱SSADT）的方法，利用產品的品質特性隨著時間衰變的現象，可以有效縮短試驗時間，並以階段性調高試驗環境應力來降低試驗樣本的個數，達到節省成本的目的。
在Tseng & Wen（2000）的文章中假設產品品質特徵值的誤差項是相互獨立的常態分佈。但是現實狀況下，產品品質特性的衰變路徑應是與時間相關的。為了克服這個缺點，本篇文章主要的研究主題是以假設產品的品質特徵值是以一逐步漂移（drift）的Wiener過程進行著，對應的產品壽命為一轉換型逆高斯分配（Transformation of Inverse Gaussian Distribution）。

本文將討論在產品的品質特徵值是以Wiener過程進行的假設下，如何進行ADT的壽命預估，及以單一應力及兩個應力進行SSADT的可靠度分析。此法不但克服了Tseng & Wen（2000）假設的不合理性，同時亦對廠商在縮短試驗時間及節省試驗成本上有所幫助。對於生產高科技產品之廠商，將大幅縮短研發產品導入市場之時間。

---

Collecting accelerated degradation test (ADT) data can provide useful lifetime information for highly reliable products if there exists a product quality characteristic whose degradation over time can be related to reliability. However, conducting a constant-stress ADT is very costly. This is obviously not applicable for assessing lifetime of a newly developed product, in which only a few test samples (prototypes) that are available. Tseng & Wen (2000) proposes a step-stress accelerated degradation test (SSADT) in this paper. A case study of LED data is used to illustrate the proposed procedure.
In their approach, the random errors are assumed to be independent and identically distributed. However, for a correlated degradation process, this assumption may be inappropriate. To overcome this difficulty, we use a Wiener Process to model a typical degradation path and statistic inference procedure for the product’s lifetime can be easily obtained. Finally, we also use an example to illustrate the difference between these two approaches.

第一章 前言

1.1 研究動機及目的 ……………….…….………………………1

1.2 文獻探討 …………………………….……………………….2

1.3 研究限制與範圍 …………………….……………………….4

1.4 研究架構 …………………………….………………….……4

第二章 衰變模型推估產品壽命之程序

2.1 Wiener過程與逆高斯分配 ………………………………… .7

2.2 衰變模型之建構 ……………………………………………..8

2.3 產品壽命的推估程序 ………………………………………..9

2.4 模擬資料分析 ……………………………….……………….10

第三章 逐步應力衰變模型及其應用

3.1 SSADT模型的應力安排 ………………………….….…….20

3.2 SSADT模型簡介 …………………………………………...21

3.3 SSADT模型之參數估計 ………………………….…..……23

第四章 模擬資料分析及敏感度分析

4.1 資料來源 ……………………………………….……………28

4.2 單一加速因子SSADT資料分析 …………….…….………29

4.3 雙變數加速因子SSADT資料分析 ….…………………….31

4.4 敏感度分析 ……………………………….………..………..33

第五章 結論與未來研究方向

5.1 結論 ……………………………………………… …………47

5.2 未來研究方向 …………………………………….. ………..48

參考文獻 49

參考文獻
[1] Boulanger, M., and Escobar, L. A., “Experimental Design for a Class of Accelerated Degradation Tests,” Technometrics, 36, 260-272 (1994).
[2] Chhikara, R. S. and Folks, L (1989), The inverse Gaussian Distribution. Theory, Methodology, and Applications, Marcel Dekker, Inc.
[3] Doksum, K. A. and Hóyland, A. (1992), “Models for Variable-Stress Accelerated Life Testing Experiments Based on Wiener Processes and the Inverse Gaussian Distribution ”, Technometrics, 34, 1, 74-82.
[4] Hoel, P. G., Port, S. C., and Stone, C. J. (1972), Introduction to Stochastic Processes, Waveland Press, Illinois 60070.
[5] Lu, C.J., and Meeker, W. Q., “Using Degradation Measures to Estimate a Time-to-Failure Distribution,” Technometrics, 35, 161-174 (1993).
[6] Meeker, W. Q., and Escobar, L. A., “A Review of Research and Current Issues in Accelerated Testing,” International Statistical Review, 61, 147-168 (1993).
[7] Nelson, W., “Analysis of Performance-Degradation Data from Accelerated Tests,” IEEE Transactions on Reliability, 30, 149-155 (1981).
[8] Nelson, W., Accelerated Testing: Statistical Models, Test Plans, and Data Analysis, John Wiley and Sons, Inc., New York, (1990).
[9] Suzuki, K., and Yokogawa, S., “An Analysis of Degradation of a Carbon Film and the Estimators,” Technical Report, The University of Electro-Communications, Tokyo, Japan, (1993).
[10] Tseng, S. T., and Chao, I. T., “Accelerated Degradation Experiment with Two Accelerated Variables,” working paper, Tsing-Hua University, 1996.
[11] Tseng, S. T., and Ku, I. H., “Optimal Screening Procedure for Highly Reliable Products”, working paper, Tsing-Hua University, 1998.
[12] Tseng, S. T.., and Wen. Z. C., “Step-Stress Accelerated Degradation Analysis For Highly Reliable Products”, Journal of Quality Technology, 32, 209-216 (2000)
[13] Yu, H. F., and Tseng, S. T., “Determining an Appropriate Termination Time for an Accelerated Degradation Experiment,” Statistica Sinica, 8(1), 207-220 (1998).

全文公開日期 本全文未授權公開 (校內網路)
全文公開日期 本全文未授權公開 (校外網路)
全文公開日期 本全文未授權公開 (國家圖書館：臺灣博碩士論文系統)