Please use this identifier to cite or link to this item:
https://scholarbank.nus.edu.sg/handle/10635/186846
Title: | ON RECURRENT EVENT DATA ANALYSIS AND DEGRADATION TEST PLANNING: NOVEL STATISTICAL METHODS IN RELIABILITY | Authors: | CAI JIAXIANG | ORCID iD: | orcid.org/0000-0001-7683-1271 | Keywords: | nonhomogeneous Poisson process, repair delay, warranty data analysis, destructive degradation test, optimal design, block effects | Issue Date: | 16-Oct-2020 | Citation: | CAI JIAXIANG (2020-10-16). ON RECURRENT EVENT DATA ANALYSIS AND DEGRADATION TEST PLANNING: NOVEL STATISTICAL METHODS IN RELIABILITY. ScholarBank@NUS Repository. | Abstract: | This thesis proposes several statistical methods to analyze reliability data with a focus on recurrent event data, and improve planning of reliability tests that generate reliability data with a focus on degradation test. In the first part, firstly, reliability data that consist of failure records are investigated. A flexible intensity-based model is proposed for repairable systems subject to repair delays. Secondly, reliability data that consist of warranty claims are considered. A statistical model that differentiates failures according to their severity and considers heterogeneity among customers is proposed. In the second part, the thesis concerns about degradation tests where product degradation has to be measured destructively. The first objective is to derive and investigate optimum plans for destructive degradation tests in closed form. The second objective is to assess block effects in the planning of accelerated destructive degradation tests, which arise from shared variability in a common test environment. | URI: | https://scholarbank.nus.edu.sg/handle/10635/186846 |
Appears in Collections: | Ph.D Theses (Open) |
Show full item record
Files in This Item:
File | Description | Size | Format | Access Settings | Version | |
---|---|---|---|---|---|---|
CaiJX.pdf | 2.47 MB | Adobe PDF | OPEN | None | View/Download |
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.