Accession Number : ADA131640

Title :   Applications of Semi-Regenerative Theory to Computations of Stationary Distributions of Markov Chains.

Descriptive Note : Technical rept.,

Corporate Author : STANFORD UNIV CA CENTER FOR RESEARCH ON ORGANIZATIONAL EFFICIENCY

Personal Author(s) : Grassmann,W K ; Taksar,Michael I

PDF Url : ADA131640

Report Date : Jan 1982

Pagination or Media Count : 27

Abstract : Arguments from Regenerative Theory have been used by a number of authors to solve equilibrium equations in queueing problems. In this paper we use Semi-Regenerative Theory, which is a generalization and sophistication of Regenerative Theroy. We believe that this is the first paper which uses Semi-Regenerative Theory for developing numerical (nonsimulation) algorithms to find the steady-state distribution of a Markov chain. The algorithm obtained is a modifications of the Gauss-Jordan method, in which all the elements used in computations are always nonnegative, which makes the algorithm numerically stable.

Descriptors :   *Queueing theory, *Markov processes, Algorithms, Stationary, Distribution theory, Theorems

Subject Categories : Operations Research

Distribution Statement : APPROVED FOR PUBLIC RELEASE