Maximum likelihood estimation and the multivariate Bernoulli distribution: An application to reliability
We investigate systems designed using redundant component configurations. If external events exist in the working environment that cause two or more components in the system to fail within the same demand period, the designed redundancy in the system can be quickly nullified. In the engineering field, such events are called common cause failures (CCFs), and are primary factors in some risk assessments. If CCFs have positive probability, but are not addressed in the analysis, the assessment may contain a gross overestimation of the system reliability. We apply a discrete, multivariate shock model for a parallel system of two or more components, allowing for positive probability that such external events can occur. The methods derived are motivated by attribute data for emergency diesel generators from various US nuclear power plants. Closed form solutions for maximum likelihood estimators exist in many cases; statistical tests and confidence intervals are discussed for the different test environments considered.
- Research Organization:
- Los Alamos National Lab., NM (United States)
- Sponsoring Organization:
- Department of Defense, Washington, DC (United States)
- DOE Contract Number:
- W-7405-ENG-36
- OSTI ID:
- 10171035
- Report Number(s):
- LA-UR-94-2275; CONF-9406224-1; ON: DE94016070
- Resource Relation:
- Conference: Lifetime data models in reliability and survival,Boston, MA (United States),14-18 Jun 1994; Other Information: PBD: [1994]
- Country of Publication:
- United States
- Language:
- English
Similar Records
A Cause-Defense Approach to the Understanding and Analysis of Common Cause Failures
Risk Analysis of Various Design Architectures for High Safety-significant Safety-related Digital Instrumentation and Control Systems of Nuclear Power Plants during Accident Scenarios