Mass casualty incidents often cause a shortage of resources for emergency medical services such as ambulances and emergency departments. These resources must be effectively managed to save as many lives as possible. Critical decisions in operating emergency medical service systems include the prioritization of patients for ambulance transport and the selection of destination hospitals. We develop a stochastic dynamic model that integrates patient transport prioritization and hospital selection problems. Policy solutions from the model are compared with other plausible heuristics, and our experimental results show that our policy solution outperforms other alternatives. More importantly, we show that there are considerable benefits from optimally selecting hospitals, which suggests that this decision is just as important as the patient prioritization decision. Motivated by the finding, we propose a heuristic policy that considers both patient prioritization and hospital selection. Experimental results demonstrate strong performance of our heuristic policy compared with existing heuristics. In addition, the proposed approach offers practical advantages. Whereas the existing heuristic policies use patient information, our heuristic policy requires information on the hospital state, which is more readily available and reliable.

大规模人员伤亡事件经常导致急救车和急诊室等紧急医疗服务的资源短缺。必须对这些资源进行有效管理，以挽救尽可能多的生命。运营紧急医疗服务系统中的关键决策包括确定患者的救护车运输优先级以及选择目的地医院。我们开发了一种随机动态模型，该模型整合了患者运输优先级和医院选择问题。将模型中的策略解决方案与其他可能的启发式方法进行了比较，我们的实验结果表明，我们的策略解决方案优于其他方法。更重要的是，我们表明，最佳选择医院有很多好处，这表明该决定与患者优先级决定同样重要。基于该发现，我们提出了一种启发式策略，该策略考虑了患者优先级和医院选择。实验结果表明，与现有启发式算法相比，我们的启发式策略具有强大的性能。另外，所提出的方法具有实际优势。现有的启发式策略使用患者信息，而我们的启发式策略需要有关医院状态的信息，该信息更容易获得和可靠。所提出的方法具有实际优势。现有的启发式策略使用患者信息，而我们的启发式策略需要有关医院状态的信息，该信息更容易获得和可靠。所提出的方法具有实际优势。现有的启发式策略使用患者信息，而我们的启发式策略需要有关医院状态的信息，该信息更容易获得和可靠。