In many practical situations, reliability data are generated in the form of counts, also called failure counts, which record the number of times that a component fails in a specified period of time. Such data may arise because of limitation of the data capture system or the way the data are reported. For example, a system may keep track of the monthly number of failures and repair them. In this paper, an attempt has been made to model such type of data using Poisson distribution in Bayesian paradigm using two different prior distributions. Moreover, for the purpose of Bayesian modelling, two important techniques, that is, approximate EM algorithm and MCMC method are implemented using R and JAGS software packages, respectively. Approximate EM algorithm is implemented using R functions for approximating the posterior densities of model parameters. Whereas, JAGS is used to approximate the posterior parameters using Gibbs sampling and the Metropolis algorithm. R and JAGS code are developed and provided. A real data set is used for the purpose of illustrations.
