Abstract

Healthcare is one of the most important disciplines to ensure life quality for human beings, especially in unpredictable scenarios such as accidents, natural disasters, or terrorism. In these cases, Emergency Medical Services (EMS) face significant challenges due to the complex nature of pre-hospital events, coordinating several resources according to the collected information in a short time. In this context, both agent-based and discrete event simulation seem to be excellent approaches to find new strategies for facing this complexity. However, there are some challenges presented in the modeling process seeking an accurate representation of the reality. In this work, we introduce a general process of pre-hospital emergency services, a hybrid simulation model based on the general process, and a machine learning approach for key simulation parameters. Our general process benefits from an intensive exchange with a multidisciplinary group of experts from medicine, including practitioners, paramedics, coordination center managers, and stakeholders, among others. Furthermore, we validated our simulation model using real-world data from an emergency coordination center from North Germany. In addition, we introduce a machine learning approach for travel speed forecasting, using seven parameters for more accuracy. Finally, we present an application of our model to analyze critical crew capacity in the coordination center.