期刊名称：Journal of Statistical Theory and Applications (JSTA)
电子版ISSN：1538-7887
出版年度：2013
卷号：12
期号：3
页码：247-254
DOI：10.2991/jsta.2013.12.3.3
语种：English
出版社：Atlantis Press
摘要：A semi-Markov process with four states, has been applied for modeling two dissimilar unit cold standby systems. At the moment that operating unit fails, the standby unit is switched to operate by using a switching device that is available with unknown probability alpha1. It is also assumed that the failure rate of unit i has the general form hi(t)= alpha2i + alpha2i+1 tbeta1-1, i=1,2, where alpha2...alpha5 are non-negative unknown parameters. In favor of semi-Markov structure of the system, maximum likelihood and the Bayes estimators of the unknown parameters alpha = (alpha1, alpha2...alpha5)are obtained while betai are non-negative known constants. Furthermore, the estimators are obtained for systems with similar units. Finally, to compare the results a simulation study is done.
关键词：Bayesian estimation;Cold standby systems;Maximum likelihood;Semi-Markov process
