Scholion.ai

Control of Stochastic Loss Networks, with Applications

Stan Zachary

Journal of the Royal Statistical Society: Series B (Statistical Methodology) · 1988 · Volume 50 · Issue 1

DOI: 10.1111/j.2517-6161.1988.tb01711.x

Abstract

We consider the application of the policy improvement routine of Markov decision theory to the control of a stochastic loss network which can process calls of differing types requiring differing resources. The objective is the minimisation of the cost of lost calls. In practice the derivation of exactly optimal policies is computationally intractable, and we identify circumstances in which nearly optimal policies may reasonably be computed. Examples are fully connected telephone networks and cellular radio networks.

Keywords

loss network, markov decision theory, policy improvement routine, dynamic channel assignment, fully connected network, cellular radio, telecommunications

Enable JavaScript for citation tools and additional paper interactions.