ROUTING WITH LIMITED STATE INFORMATION IN QUEUING-SYSTEMS WITH BLOCKING

We prove the optimality of routing policies with limited state informa tion that are employed in queueing systems with finite capacities and customers arriving at arbitrary instants. Using sample path arguments we show that, on one hand, when service times are exponential and capa cities are equal, the round robin policy minimizes the expected number of losses by any time t and at the same time, maximizes the expected number of departures by t, over all policies that use no state informa tion. On the other hand, when service times are deterministic, a modif ied round robin policy that makes use of limited state information out performs stochastically all dynamic policies that have richer state in formation, in terms of the number of losses and the number of departur es by t.