This thesis is concerned with reliability analysis of periodic switching standby systems where standby units can fail. Large equipments, such as power plants, are usually mechanical and designed as a standby system to increase reliability. Mechanical equipments fail mainly due to degradation. Degradation can cause failures of standby units but failures can not be detected in standby state. In such situations, it is a common practice to perform periodic switching to detect failures during standby state.
Cumulative exposure model is introduced to represent the lifetime distribution of standby systems. Method of obtaining maximum likelihood estimators is outlined, their asymptotic properties are studied and specific formula for Weibull distribution are obtained when the number of standby units is . A simulation procedure is proposed for assessing the reliability of standby systems and obtaining optimal switching periods. Simulation studies are performed to investigate the effects of switching failures and number of standby units.
