A formal proof for the insensitivity of simple bounds for finite multi- server non-exponential tandem queues based on monotonicity results (Q1100815): Difference between revisions

For a model of two multiservers in tandem involving the possibility of arriving calls getting lost, the author obtains a lower and an upper bound for the steady-state probability of such a loss. These bounds are themselves loss probabilities, obtained for modifications of the model. The modifications are chosen in such a way that the corresponding loss probabilities are insensitive in the sense that they depend on the service time distributions only through their means. It is shown that their property as bounds is insensitive in the same sense.