Approximation Methods for the Minimum Average Cost Per Unit Time Problem with a Diffusion Model.

Abstract

Approximation methods for the minimum average cost per unit time problem with a controlled diffusion model is treated. In order to work with a bounded state space, the reflecting diffusion model of Strook and Varadhan is used, although other models can also be treated. The control problem is approximated by an average cost per unit time problem for a Markov chain, and weak convergence methods are used to show convergence of the minimum costs to that for the optimal diffusion. The procedure is quite natural and allows the approximation of many interesting functionals of the optimal process.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1978
Accession Number
ADA058876

Entities

People

  • Harold J. Kushner

Organizations

  • Brown University

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Air Force
  • Applied Mathematics
  • Boundaries
  • Computational Science
  • Computer Programming
  • Convergence
  • Diffusion
  • Dynamic Programming
  • Equations
  • Interpolation
  • Markov Chains
  • Mathematics
  • Probability
  • Sequences
  • Stationary
  • Stationary Processes
  • Weak Convergence

Fields of Study

  • Mathematics

Readers

  • Mathematical Modeling and Probability Theory.

Technology Areas

  • Space