Models estimate ground and helicopter EMS travel times

The study evaluated methods to estimate emergency medical service travel times to incidents and hospitals, with the aim of helping dispatchers compare ground and helicopter response. Ground travel estimates from an open-source routing tool were adjusted using historical travel data with linear regression and gradient boosting. Helicopter estimates were based on great-circle distances adjusted using historical data.

In validation data collected after the methods were finalized, the best-performing ground models explained 79% of variation in drive times to incidents and 70% in drive times to hospitals (R² 0.79 and 0.70). Helicopter models explained less variation in flight times to incidents and hospitals (R² 0.46 and 0.38). Mean absolute percentage error was 18–20% across the estimation techniques and time intervals.

The authors say tools based on the study’s open-source code could help dispatchers estimate relative travel times and assess potential time benefits when choosing between resources. The reported prediction errors and lower performance for helicopter estimates are important limitations; the study evaluated travel-time prediction, not patient outcomes or whether model-assisted dispatch improves care.