Forecasting saturation in the Emergency Department: a comparison of queuing data-driven approaches

Emergency Departments are highly stochastic environments caused by the unscheduled patients’ arrivals and the diverse nature of their health problems. Health services research, statistical modeling and operational research all attempt to give tools and strategies to manage crowding through crowding scores, forecasting and optimization models. This paper proposes a novel approach that evaluates crowding using a new saturation indicator. It also sets an original data-driven queuing network to model the system and help forecasting crowding and then optimizes it. This study is illustrated using data from the Emergency Department of Troyes hospital.