A Markov decision process model of allocating emergency medical resource among multi-priority injuries

Abstract

Abstract:This article studies how to allocate a scare emergency medical resource at the beginning of large-scale disaster. One salient problem in emergency rescue is how to coordinate the resource among injuries of different degrees in different periods. The injuries are classified into three levels and given different priorities. Waiting cost with resources shortage, deteriorated cost with delay and transferring cost to other hospitals are introduced into the decision-making process. Basing on incorporating the costs into revenue function, a Markov decision process (MDP) model is proposed to establish the optimal action policies in different periods. A dynamic algorithm is proposed to derive the optimal solution. With a numerical experiment, the improvement of MDP is displayed and some managerial suggestions are also provided.Keywords: emergency management; medical resource allocation; Markov decision process; multi-priority injuries; operational research.

A Visual Analysis of Frontiers and Dynamic Tendencies for Humanitarian Logistics

“…Rennemo et al [37] used a function of linear utility to search humanitarian aid distribution equity in a three-stage stochastic programming model for disaster response planning. Ni and Zhao [38] presented a Markov decision model to maximize lifesaving against an emergency medical resource allocation problem. In addition, commercial software has been updated and embedded with some common algorithms, which has become a powerful optimization platform.…”

Optimizing Daily Service Scheduling for Medical Diagnostic Equipment Considering Patient Satisfaction and Hospital Revenue

“…As mentioned above, since the study is based on the schedule on the service day, it is necessary to construct the state of patients at the next time (n + 1) in terms of whether the appointment is made and whether the appointment is delayed. Therefore, in the next time state representation model of MDP patients [36,[38][39][40][41][42][43][44][45], two additional binary variables Z B and Z C are added to explain the delayed arrival of patients on appointments.…”