Optimal Control of a Brownian Storage System

Abstract

Consider a storage system (such as an inventory or bank account) whose content fluctuates as a Brownian Motion X = (X(t), t> or = 0) in the absence of any control. Let Y = (Y(t), t > or = 0) and Z = (Z(T), t > or = 0) by non-decreasing, non-anticipating functionals representing the cumulative input to the system and cumulative output from the system respectively. The problem is to choose Y and Z so as to maximize expected discounted reward subject to the requirement that X(t) + Y(t) - Z(t) > or = 0 for all t > or = 0 almost surely. In our first formulation, we assume that a reward of one dollar is received for every unit of output, while a cost of k > 1 dollars is incurred for every unit of input. We explicitly compute an optimal policy involving a single critical number. In our second formulation, the cumulative input Y is required to be a step function, and an additional cost of K > 0 dollars is incurred each time that an input jump occurs. We explicitly compute an optimal policy involving two critical numbers. Applications to inventory/production control and stochastic cash management are discussed.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1976
Accession Number
ADA030650

Entities

People

  • Allison J. Taylor
  • J. M. Harrison

Organizations

  • Stanford University

Tags

DTIC Thesaurus Topics

  • Boundaries
  • Brownian Motion
  • Classification
  • Differential Equations
  • Diffusion
  • Equations
  • Inventory
  • Inventory Control
  • Markov Processes
  • Materials
  • Production
  • Production Control
  • Random Variables
  • Step Functions
  • Stochastic Control
  • Stochastic Processes
  • United States

Readers

  • Computer Science.
  • Government Contracting/Procurement.
  • Statistical inference.