Predicting overcrowding in the acute care domain using random forest regression

Crowding is a phenomenon that occurs more frequently within the facilities of the acute care domain. Acute care providers are interested in; means to quantify crowding, potential predictors of crowding, and a model to forecast crowding. This information can help the facilities to plan accordingly to future acute care demand. In this research, we focus on these topics and provide a machine learning model for an emergency department and general practitioner's post in Gelderland. This model predicts the total daily visitors to the facilities based on a variety of external variables, such as weather data, pollen data, et cetera and provides an overview of the most important variables in the models.