Abstract—Emergency Medical Services (EMS), play a vital role for community well-being, provides lifesaving assistance during emergencies. Accurately forecasting emergency call demand is crucial for optimizing resource allocation and improving response times. In this study, we analyzed an EMS dataset containing emergency call details from four U.S. states to develop a predictive model. We utilized the Autoregressive Integrated Moving Average (ARIMA) model, a widely adopted method for analyzing and forecasting stationary time series data. To fine-tune the ARIMA model’s hyperparameters, we implemented three methods: Auto-ARIMA, grid search, and Bayesian Optimization (BO). Although Auto-ARIMA and grid search generated reasonable predictions, BO yielded superior accuracy with more precise forecasts. This finding underscores the superiority of BO for time series prediction tasks. The finding of this study could help EMS organizations in effectively predicting demand, leading to better resource allocation, enhanced operational efficiency, and faster response times. Additionally, these precise predictions can strengthen EMS systems’ capacity to handle emergencies and improve overall health infrastructure.
 
Keywords—Emergency Medical Services (EMS), Bayesian Optimization (BO), Autoregressive Integrated Moving Average (ARIMA) model


Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).