Optimizing a UAV-based Emergency Medical Service Network for Trauma Injury Patients

Emergency medical service must be time sensitive. However, in many cases, satisfactory service cannot be ensured due to inconvenient logistics. For its easily deployable and widely accessible nature, unmanned aerial vehicles (UAVs) have the potential to improve the service, especially in areas that are traditionally under-served. In this paper, we develop a service network optimization problem for locating UAV bases, staffing a UAV fleet at each constructed base, and zoning demand nodes. We formulate a location-allocation optimization model with numerically simulated waiting times for the service zones as the objective. We adapt a genetic algorithm to solve the optimization model. We test our network optimization approach on instances of traumatic injury cases. By comparing our approach to a two-phase method in Boutilier et al. [1], we suggest an up to 60% reduction in mean waiting time.