In this paper, a method is presented for the efficient estimation of rare-event (overflow) probabilities in Jackson queueing networks using importance sampling. The method differs in two ways from methods discussed in most earlier literature: the change of measure is state-dependent, i.e., it is a function of the content of the buffers, and the change of measure is determined using a cross-entropy-based adaptive procedure. This method yields asymptotically efficient estimation of overflow probabilities of queueing models for which it has been shown that methods using a stateindependent change of measure are not asymptotically efficient. Numerical results demonstrating the effectiveness of the method are presented as well.
|Title of host publication||The 2000 Winter Simulation Conference (WSC'00)|
|Place of Publication||Orlando, Florida, USA|
|Number of pages||10|
|Publication status||Published - Dec 2000|