Resource allocation for reliability of a complex system with aging components
- Los Alamos National Laboratory
To assess the reliability of a complex system, many different types of data may be available. Full-system tests are the most direct measure of reliability, but may be prohibitively expensive or difficult to obtain. Other less direct measures, such as component or section level tests, may be cheaper to obtain and more readily available. Using a single Bayesian analysis, multiple sources of data can be combined to give component and system reliability estimates. Resource allocation looks to develop methods to predict which new data would most improve the precision of the estimate of system reliability, in order to maximally improve understanding. In this paper, we consider a relatively simple system with different types of data from the components and system. We present a methodology for assessing the relative improvement in system reliability estimation for additional data from the various types. Various metrics for comparing improvement and a response surface approach to modeling the relationship between improvement and the additional data are presented.
- Research Organization:
- Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- AC52-06NA25396
- OSTI ID:
- 960610
- Report Number(s):
- LA-UR-08-05483; LA-UR-08-5483; QREIE5; TRN: US201006%%1250
- Journal Information:
- Quality and Reliability Engineering International, Journal Name: Quality and Reliability Engineering International; ISSN 0748-8017
- Country of Publication:
- United States
- Language:
- English
Similar Records
Identification of components to optimize improvement in system reliability
Assessing system reliability and allocating resources: a bayesian approach that integrates multi-level data