Abstract:
In this paper a two-unit intermittently used cold standby system is considered, introducing a model where failure of a unit is detected whenever it occurs, regardless of the state of the system (need or no-need). There is a single repair facility, and repair of a unit will start at the epoch of a failure, provided that the repair facility is available. This system allows that n observations are taken at widely spaced points in time and that only the state that prevails at the instant of sampling is revealed. The observations in this case are called snapshots. A so-called monitoring device can then be attached to the system which observe the different states in the process. Bayesian analysis of the performance variable, namely stationary rate of disappointments, is introduced; assuming an informative prior for the unknown parameters in the model. Gibbs sampling is used to derive the posterior distribution of the stationary rate of disappointments and subsequently the highest posterior density (HPD) interval. A numerical example illustrates the theory.
Page(s):
223-229
DOI:
DOI not available
Published:
Journal: Pakistan Journal of Statistics, Volume: 20, Issue: 2, Year: 2004