NSF Supported Research Advances Data Driven Decision Making for Emergency Medical Services
A multi-institution collaboration involving the USC Sonny Astani Department of Civil and Environmental Engineering, Columbia University, and the New York City Fire Department is developing next generation digital tools to improve the speed and reliability of emergency medical services. Supported by a National Science Foundation Smart and Connected Communities award, the project focuses on creating probabilistic simulation models that help EMS agencies evaluate new operational policies in a safe virtual environment.
The research is led by USC Assistant Professor Audrey Olivier, whose work integrates physics-based modeling, data analytics, and uncertainty quantification to support resilient urban infrastructure. The team aims to design systems that can better manage the inherent uncertainty of 911 operations, including fluctuating call volumes, variable traffic conditions, hospital bottlenecks, and potential gaps in ambulance coverage.
At the core of the effort is the development of a digital shadow that can run thousands of simulated scenarios to test policies and predict cascading impacts across the EMS system. Unlike fully integrated real-time digital twins, the digital shadow provides an offline environment for agencies to explore new dispatch strategies, risk-averse decision making methods, and rare-event scenarios that traditional rule-based models often overlook.
The researchers are designing tools that quantify uncertainty, simulate high-impact but infrequent failures, and incorporate real-time data such as traffic patterns. Their goal is to support decisions that improve performance during both routine operations and critical high-risk events when delays carry significant consequences.
The partnership builds on earlier collaborations between Olivier’s team and FDNY, which produced algorithms that reduced average hospital transport times and helped balance patient loads across New York City hospitals. The new project expands this work to consider the full EMS workflow, integrating dispatch, routing, hospital assignment, and systemwide coordination.
Researchers note that the work extends beyond EMS. The methods being developed are broadly applicable to cyber-physical systems where human decisions interact with dynamic infrastructure. Insights from this project could support resilience in power systems, transportation networks, and other urban systems that must operate under uncertainty.
By bringing together academic researchers and emergency response practitioners, the project aims to generate actionable, scalable policies that enhance the safety and reliability of emergency medical response for one of the country’s largest urban EMS agencies.
Read More: Every Second Counts: Leveraging Data to Streamline Emergency Medical Services