<p>Platelets over-ordering by hospitals is problematic and causes unnecessary wastage and costs of storage and purchase on one side and cost of shortage on the other side. This paper pioneers the design of a dynamic and intelligent scheme for platelets inventory-distribution policy which addresses the over-ordering concern using information about hospitals’ demand, orders, and their honesty in placing orders. Hospitals’ honesty affects the priority of hospitals to receive their orders. Employing artificial intelligence, the hospitals’ performance is analyzed by an intelligent agent in a deep reinforcement learning framework to make purchase and distribution decisions adaptable to the uncertain and dynamic environment. The dynamic mechanism modifies orders, transfers and priorities, and contributes to the fair distribution of platelets among hospitals. Besides several test instances, the practical application of the proposed framework is substantiated through a real topology. The computational results of the proposed scheme highlight cost savings of up to 13.68% and 64.10% compared to two approaches taken from the literature as benchmarks.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

A dynamic and intelligent decision-making framework for a platelet inventory-distribution network

  • Sara Cheraghi,
  • Abdorrahman Haeri,
  • Seyed Farid Ghannadpour

摘要

Platelets over-ordering by hospitals is problematic and causes unnecessary wastage and costs of storage and purchase on one side and cost of shortage on the other side. This paper pioneers the design of a dynamic and intelligent scheme for platelets inventory-distribution policy which addresses the over-ordering concern using information about hospitals’ demand, orders, and their honesty in placing orders. Hospitals’ honesty affects the priority of hospitals to receive their orders. Employing artificial intelligence, the hospitals’ performance is analyzed by an intelligent agent in a deep reinforcement learning framework to make purchase and distribution decisions adaptable to the uncertain and dynamic environment. The dynamic mechanism modifies orders, transfers and priorities, and contributes to the fair distribution of platelets among hospitals. Besides several test instances, the practical application of the proposed framework is substantiated through a real topology. The computational results of the proposed scheme highlight cost savings of up to 13.68% and 64.10% compared to two approaches taken from the literature as benchmarks.