Boston Children's Hospital deployed predictive AI models that forecast emergency department admissions and hospital bed availability, helping clinical staff plan capacity for elective procedures and respond to surges in patient demand. During the 'tripledemic' of RSV, COVID-19, and flu, the models accurately predicted bed availability.

The hospital implemented predictive modeling for emergency hospital admissions and integrated disease surveillance data for capacity planning purposes, as confirmed by the Northeastern University Institute for Experiential AI. The hospital also piloted a tool called POPP (Prediction of Patient Placement), a real-time forecasting tool that predicts the likelihood of ED admission within 10 to 20 minutes of patient triage, transforming a static predictive model into a real-time capacity dashboard for clinical staff. Current status of the POPP pilot beyond 2021 reporting is unclear from available sources.