Estimating the change point of Erlang interarrival time distribution

This paper discusses the problem of the change point for the interarrival time distribution in the context of exponential families for the Ek/G/c queueing system. Bayes estimates of the posterior probabilities and the positions of change are derived from the Erlang distribution. Numerical examples are simulated using Monte Carlo techniques for various sample sizes of the arriving customers at different positions of change based on the Erlang distribution. It is observed that the estimates are precise and very close to their true values.