Abstract
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 ……………….…….………………………11.2 文獻探討 …………………………….……………………….21.3 研究限制與範圍 …………………….……………………….41.4 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 ………………………………… .72.2 衰變模型之建構 ……………………………………………..82.3 產品壽命的推估程序 ………………………………………..92.4 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 ………………………….….…….203.2 SSADT模型簡介 …………………………………………...213.3 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 ……………………………………….……………284.2 單一加速因子SSADT資料分析 …………….…….………294.3 雙變數加速因子SSADT資料分析 ….…………………….314.4 敏感度分析 ……………………………….………..………..33第五章 結論與未來研究方向5.1 結論 ……………………………………………… …………475.2 未來研究方向 …………………………………….. ………..48參考文獻 49