Optimal Customer Selection in Exponential Queues.

Abstract

The paper discusses the problem of maximizing the long-run average return of a facility with M servers, a queue of capacity N and Poisson arrivals at a rate mu, by selecting the customer upon arrival (case alpha) or upon entry to the service (case beta) according to his offer of a reward R and of an exponential service rate lambda determined by a joint distribution. After formulating the system as a semi-Markov decision process, the author establishes simple recurrence relationships on the critical quantities characterizing the optimal stationary policy pi prime and discusses its structure. (Author)

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1971
Accession Number
AD0732496

Entities

People

  • Michel Cramer

Organizations

  • University of California, Berkeley

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Stationary

Readers

  • Mathematical Modeling and Probability Theory.
  • Regression Analysis.