{"title":"A hybrid approach for modeling degradation processes with a degradation free stage","authors":"R. Jiang","doi":"10.1109/ICRSE.2017.8030734","DOIUrl":null,"url":null,"abstract":"The delay time model is typically used to optimize an inspection scheme. A challenging issue with this model is that the exact defect initiation time and failure time are often not available. However, it is possible to build a delay time model for the situation where the degradation process has a degradation free stage. In this paper, we develop a hybrid approach to model a delay time model based on a set of incomplete data. The degradation data considered in this paper are obtained through destructive tests. The sample size is small and information is incomplete. A two-step modeling approach is proposed to model the data. In the first step we estimate the degradation initiation time distribution using mean pseudo-initiation time and implicit censoring information; and in the second stage we estimate the failure time distribution based on a non-homogeneous normal process model with a power-law mean value function. Application of the resulting model is also discussed.","PeriodicalId":317626,"journal":{"name":"2017 Second International Conference on Reliability Systems Engineering (ICRSE)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 Second International Conference on Reliability Systems Engineering (ICRSE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICRSE.2017.8030734","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The delay time model is typically used to optimize an inspection scheme. A challenging issue with this model is that the exact defect initiation time and failure time are often not available. However, it is possible to build a delay time model for the situation where the degradation process has a degradation free stage. In this paper, we develop a hybrid approach to model a delay time model based on a set of incomplete data. The degradation data considered in this paper are obtained through destructive tests. The sample size is small and information is incomplete. A two-step modeling approach is proposed to model the data. In the first step we estimate the degradation initiation time distribution using mean pseudo-initiation time and implicit censoring information; and in the second stage we estimate the failure time distribution based on a non-homogeneous normal process model with a power-law mean value function. Application of the resulting model is also discussed.