Basic Research in Reliability for Real Systems
Annual technical rept. 1 Aug 1988-31 Jul 1989
UNIVERSITY OF SOUTHERN CALIFORNIA LOS ANGELES DEPT OF ELECTRICAL ENGINEERING
Pagination or Media Count:
The goal of our research is to develop practical models and efficient algorithms to analyze and evaluate the reliabilityavailabilitymaintainability of complex systems in which component failures are statistically dependent and each component is subject to degradations before complete failure. We have developed the Event-Based Reliability Model EBRM for the reliability modeling and analysis of real systems in which component failures are statistically dependent. Most existing reliability models assume that system component failures are statistically independent. This assumption, though it greatly simplifies the problem, is often not valid, and the result is usually an overestimation of network reliability. We have also developed a model to approximate the reliability of systems with multimode components. Previous research on reliability has been focused on models which assume that each component may be in one of two modes, namely, operative or failed. In real life, a component may undergo degradations in performance before a complete outage, and will therefore operate in more than two modes. More recently, we have developed the Cause-based Multimode Model CBMM, which allows one to consider failure dependencies of components which are subject to degradations.
- Statistics and Probability