串聯系統存在隱蔽資料之可靠度分析─以廣義伽瑪分配為例; The relibility analysis of series system with masked data dash generalized gamma distribution

NCU Institutional Repository > 理學院 > 統計研究所 > 博碩士論文 > Item 987654321/48262

請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/48262

題名:	串聯系統存在隱蔽資料之可靠度分析─以廣義伽瑪分配為例;The relibility analysis of series system with masked data dash generalized gamma distribution
作者:	蘇岏智;WanJ-Jr Su
貢獻者:	統計研究所
關鍵詞:	廣義伽瑪分配;貝氏因子;期望值-最大化演算法;馬卡夫鏈蒙地卡羅演算法;隱蔽資料;Markov chain Monte Carlo;EM algorithm;bayesian factor;DIC;masked data;generalized gamma
日期:	2011-07-15
上傳時間:	2012-01-05 14:42:51 (UTC+8)
摘要:	在串聯系統中，當任一物件失效即導致系統停止運作，但有時某些因素導致引發系統失效之物件無從觀測，亦即資料為隱蔽資料。本文討論在不同隱蔽水準時，物件壽命分別具有韋伯分配與廣義伽瑪分配的可靠度之試驗。我們以期望值最大化演算法求得模型中參數之最大概似估計和以無母數拔靴法估計參數、可靠度和壽命的標準誤；並在主觀先驗分佈下由馬可夫鍊蒙地卡羅方法得貝氏估計，同時比較兩種方法在物件與系統之平均壽命及可靠度函數之統計推論。並以傳統的概似比檢定、AIC 和 BIC 方法與貝氏選模中常用的 DIC 和貝氏因子法則探討資料配適廣義伽瑪分配與韋柏分配的模型選擇問題。模擬結果顯示，當樣本資訊不足時，貝氏分析所得結果優於最大概似方法。 In this thesis, we consider a system of independent and non-identical components connected in series, each component having a Weibull life time distribution under Type-I censored. In a series system, the system fails if any of the components fails, and it may only be ascertained that the cause of system failure is due to one of the components in some subset of system components, so called masked data. The maximum likelihood estimates via EM algorithm is developed for the model parameters with the aid of nonparametric bootstrap method to estimate the resulting standard errors of the MLE when the data are masked. Subjective Bayesian inference incorporated with the Markov chain Monte Carlo method is also addressed. Simulation study shows that the Bayesian analysis provides better results than the maximum likelihood approach not only in parameters estimation but also in reliability inference for both the system and components. We also discuss model fitting issue regarding the generalized gamma distribution and Weibull distribution via different model selection criteria.
顯示於類別:	[統計研究所] 博碩士論文

文件中的檔案:

檔案	描述	大小	格式	瀏覽次數
index.html		0Kb	HTML	706	檢視/開啟

在NCUIR中所有的資料項目都受到原著作權保護.

社群 sharing

資料載入中.....