MATH MODELING

On the application of machine learning methods to solving problems queuing theory

V. M. Vishnevsky, A. V. Gorbunova

V. A. Trapeznikov Institute of Control Sciences of Russian Academy of Sciences, Moscow, Russia

Abstract: In this paper, for the first time, a systematic presentation of a new method for studying queuing systems (QS) using machine learning algorithms is presented. The publications considered in the review are divided into several categories – articles in which machine learning algorithms are used to predict the parameters of the QS of a technical nature and publications in which machine learning is used to estimate the probabilistic-temporal characteristics of the QS. The analysis of publications allows us to conclude that the application of machine learning methods is highly effective, the prospects for further research, as well as the possible separation of a new approach into an independent direction in the field of solving complex problems of queuing theory.

Keywords: queuing theory, queuing network, queuing system, fork-join queue, simulation, data mining, machine learning, artificial neural networks.

DOI: 10.14357/20718632210407

Bibliographic databases:

•