On the Distribution of the Number Stranded in Bulk-arrival, Bulk-service Queues of the M/G/1 Form

Abstract

Bulk-arrival queues with single servers that provide bulk service are widespread in the real world, e.g.; elevators in buildings, people-movers in amusement parks, air-cargo delivery planes, and automated guided vehicles. Much of the literature on this topic focusses on the development of the theory for waiting time and number in such queues. We develop the theory for the number stranded, i.e.; the number of customers left behind after each service, in queues of the M/G/1 form, where there is single server, the arrival process is Poisson, the service is of a bulk nature, and the service time is a random variable. for the homogenous Poisson case, in our model the service time can have any given distribution. for the non-homogenous Poisson arrivals, due to a technicality, we assume that the service time is a discrete random variable. Our analysis is not only useful for performance analysis of bulk queues but also in designing server capacity when the aim is to reduce the frequency of stranding. Past attempts in the literature to study this problem have been hindered by the use of Laplace transforms, which pose severe numerical difficulties. Our approach is based on using a discrete-time Markov chain, which bypasses the need for Laplace transforms and is numerically tractable. We perform an extensive numerical analysis of our models to demonstrate their usefulness. to the best of our knowledge, this is the first attempt in the literature to study this problem in a comprehensive manner providing numerical solutions. © 2011 Elsevier B.V. All rights reserved.

Department(s)

Engineering Management and Systems Engineering

Keywords and Phrases

Bulk queues; Downside risk; Queueing

International Standard Serial Number (ISSN)

0377-2217

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2024 Elsevier, All rights reserved.

Publication Date

16 Jul 2011

Share

 
COinS