Please use this identifier to cite or link to this item: https://hdl.handle.net/11264/921
Title: New and Extended Results in Renewal and Queueing Theories
Authors: Kim, James
Royal Military College of Canada / Collège militaire royal du Canada
Chaudhry, Mohan
Keywords: Stochastic processes, Markov chains, Renewal theory, Discrete-time, Bulk-renewal processes, Queueing theory, Single-server queues, Service stages.
Issue Date: 3-May-2016
Abstract: This thesis encompasses new and extended results in renewal and queueing theories. In the renewal theory portion of this thesis, the asymptotic result of renewal mass function and new asymptotic moments are found using the method of generating functions. This method is not only simple but also provides the extra constant terms in the asymptotic second moment which are unavailable in the literature. Higher asymptotic moments and their corresponding extra constant terms can also be found using the method of generating functions. Previous results in the existing literature do not have these extra constant terms. Recent work in renewal theory has the extra constant terms in a non-bulk renewal processes. The purpose of this thesis is to extend that recent work to the bulk-renewal processes in discrete-time. In the queueing theory portion of this thesis, the imbedded Markov chain technique is used to determine the distributions of the number of uncompleted service stages, the number of customers in the system, and the waiting-time-in-queue. Single-server queues with a fixed number of service stages have been analyzed by many authors, some of whom state that there is no simple way to analyze the queue 𝐺𝐼/𝐸𝑋/1. The purpose of this thesis is to review and extend the previous work on 𝐺𝐼/𝐸𝑟/1 to the more general model 𝐺𝐼/𝐸𝑋/1 in which the number of stages is randomly distributed.
URI: https://hdl.handle.net/11264/921
Appears in Collections:Theses

Files in This Item:
File Description SizeFormat 
draft 110 (29 April 16) (1).pdfMain article1.55 MBAdobe PDFThumbnail
View/Open


Items in eSpace are protected by copyright, with all rights reserved, unless otherwise indicated.