TY - GEN

T1 - Can Machines Solve General Queueing Problems?

AU - Baron, Opher

AU - Krass, Dmitry

AU - Sherzer, Eliran

AU - Senderovich, Arik

N1 - Publisher Copyright:
© 2022 IEEE.

PY - 2022

Y1 - 2022

N2 - We study how well a machine can solve a general problem in queueing theory, using a neural net to predict the stationary queue-length distribution of an M/G/1 queue. This problem is, arguably, the most general queuing problem for which an analytical 'ground truth' solution exists. We overcome two key challenges: (1) generating training data that provide 'diverse' service time distributions, and (2) providing continuous service distributions as input to the neural net. To overcome (1), we develop an algorithm to sample phase-type service time distributions that cover a broad space of non-negative distributions; exact solutions of M / PH /1 (with phase-type service) are used for the training data. For (2) we find that using only the first n moments of the service times as inputs is sufficient to train the neural net. Our empirical results indicate that neural nets can estimate the stationary behavior of the M/G/1 extremely accurately.

AB - We study how well a machine can solve a general problem in queueing theory, using a neural net to predict the stationary queue-length distribution of an M/G/1 queue. This problem is, arguably, the most general queuing problem for which an analytical 'ground truth' solution exists. We overcome two key challenges: (1) generating training data that provide 'diverse' service time distributions, and (2) providing continuous service distributions as input to the neural net. To overcome (1), we develop an algorithm to sample phase-type service time distributions that cover a broad space of non-negative distributions; exact solutions of M / PH /1 (with phase-type service) are used for the training data. For (2) we find that using only the first n moments of the service times as inputs is sufficient to train the neural net. Our empirical results indicate that neural nets can estimate the stationary behavior of the M/G/1 extremely accurately.

UR - http://www.scopus.com/inward/record.url?scp=85147441076&partnerID=8YFLogxK

U2 - 10.1109/WSC57314.2022.10015451

DO - 10.1109/WSC57314.2022.10015451

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85147441076

T3 - Proceedings - Winter Simulation Conference

SP - 2830

EP - 2841

BT - Proceedings of the 2022 Winter Simulation Conference, WSC 2022

A2 - Feng, B.

A2 - Pedrielli, G.

A2 - Peng, Y.

A2 - Shashaani, S.

A2 - Song, E.

A2 - Corlu, C.G.

A2 - Lee, L.H.

A2 - Chew, E.P.

A2 - Roeder, T.

A2 - Lendermann, P.

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2022 Winter Simulation Conference, WSC 2022

Y2 - 11 December 2022 through 14 December 2022

ER -