Prediction of the optimal control in a multi-server heterogeneous queueing system

In this paper we consider a multi-server heterogeneous queueing system where servers differ in service rates and operating costs. The optimal allocation policy for this system is of threshold type. Gathering data for evaluated optimal threshold levels and system parameters is performed using a policy iteration algorithm. We study the possibility to use these data-sets to provide predictions for optimal thresholds through artificial neural networks. The obtained results are accompanied by heuristic solution based on a uid approximation. Numerical examples illustrate the quality of provided predictions. Copyright © 2020 for this paper by its authors.