Analysis of blocking networks with temporal dependence

In this paper we estend the class of MAP queueing networks to include blocking models. We consider two different blocking mechanisms: Repetitive Service-Random Destination and Blocking After Service. We analyze the Markov process underlying the MAP queueing network and propose a methodology based on a partition of the state space into “marginal state spaces”. By using this partition, we prove a set of “partial” balance equations that relates blocking performance indexes. The proposed methodology can be a sound framework to define approximate solution methods for MAP queueing networks with blocking.