A markov model for inventory level optimization in supply-chain management

  • Authors:
  • Scott Buffett

  • Affiliations:
  • Institute for Information Technology – e-Business, National Research Council Canada, Fredericton, New Brunswick, Canada

  • Venue:
  • AI'05 Proceedings of the 18th Canadian Society conference on Advances in Artificial Intelligence
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose a technique for use in supply-chain management that assists the decision-making process for purchases of direct goods Based on projections for future prices and demand, requests-for-quotes are constructed and quotes are accepted that optimize the level of inventory each day, while minimizing total cost The problem is modeled as a Markov decision process (MDP), which allows for the computation of the utility of actions to be based on the utilities of consequential future states Dynamic programming is then used to determine the optimal quote requests and accepts at each state in the MDP The model is then used to formalize the subproblem of determining optimal request quantities, yielding a technique that is shown experimentally to outperform a standard technique from the literature The implementation of our entry in the Trading Agent Competition-Supply Chain Management game is also discussed.