Analysis of persistence of relevance in systems with imperfect fault coverage

3Citations
Citations of this article
5Readers
Mendeley users who have this article in their library.
Get full text

Abstract

This paper introduces the concept of persistence and analyzes its influence on reliability of systems with imperfect fault coverage (IFC). A component is persistent if it is always relevant in the system unless the system is failed, and a system is persistent if all the components are persistent. In traditional imperfect fault coverage models, simply coverage models (CMs), the coverage (including identification and isolation) is typically limited to the faulty components regardless of their relevance. The general assumption on system coherence cannot guarantee that a component will be always relevant in the system. Rather, an initially relevant component could become irrelevant later due to the failures of other components, which is unfortunately a missing issue in the traditional CMs. For systems with IFC, it is important to cover the non-persistent components whenever they become irrelevant, so as to prevent their future uncovered faults that may lead to system failure. A new coverage model incorporating the timely coverage of the irrelevant components (in addition to the faulty components) is proposed, which opens up a cost-effective approach to improve the system reliability without increasing redundance. © 2014 Springer International Publishing.

Cite

CITATION STYLE

APA

Xiang, J., Machida, F., Tadano, K., & Maeno, Y. (2014). Analysis of persistence of relevance in systems with imperfect fault coverage. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8666 LNCS, pp. 109–124). Springer Verlag. https://doi.org/10.1007/978-3-319-10506-2_8

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free