Reliability, Yield, and Stress Burn-in: A Unified Approach for Microelectronics Systems Manufacturing and Software Development
Building a reliable P2P system out of unreliable P2P clients: the case of KAD
CoNEXT '07 Proceedings of the 2007 ACM CoNEXT conference
| Hi-index | 0.00 |
In this paper, we introduce and analyze simple, new inspection policies for maintaining deteriorating equipment with non-self-announcing failures. These policies utilize information from the inspection/repair history as well as the system lifetime distribution to schedule future inspections. Numerical results indicate that the policies can significantly outperform standard (periodic) maintenance strategies by providing higher availability for a given inspection rate.