Introduction
This article introduces a novel emergency response optimization framework that integrates simulation with optimization to improve the efficiency of emergency response to out-of-hospital cardiac arrests (OHCA) in urban environments.
OHCA is a critical medical emergency, and timely response is essential for patient survival. Traditional emergency response systems face challenges in reaching patients within the critical four-minute window due to complex urban road networks and uncertain incident locations. This study proposes an integrated simulation-based framework that combines GIS and agent-based modeling using AnyLogic software to address these challenges.
Simulation model
Researchers developed an agent-based simulation model using AnyLogic to optimize emergency response to out-of-hospital cardiac arrest (OHCA) incidents in urban environments.
The model simulates out-of-hospital cardiac arrest events and the corresponding actions of responders, including CPR (Cardiopulmonary Resuscitation), AED delivery, and AED operation. Each agent represents either a patient or a responder, with specific attributes such as mobility, skill level, and response behavior. The model incorporates real-world geographic data to simulate responder movements through road networks and urban environments, allowing for more accurate prediction of response times and survival outcomes.
The simulation framework combines GIS integration and event probability models with agent-based simulation to create a detailed and flexible emergency response system. It evaluates the effectiveness of different dispatch strategies, responder deployment numbers, and skill compositions.
Key features include GIS-based grids to replicate city layouts and road networks, as well as agent state charts to simulate the progression of OHCA incidents from the occurrence of the event to the arrival of responders.
Agents transition between states such as "OHCA occurrence", "Responder Dispatch", "Responder Arrival", and "Patient Outcome", with probabilities based on response times. The model allows customization of dispatch ranges, responder types, and skill ratios to visualize the impact on survival rates and response efficiency over time.
Read also: Explore additional resources on healthcare simulation.
Results
The proposed emergency response simulation was tested in Shenzhen city, focusing on the effects of dispatch range, the number of responders, and the ratio of skilled to mobile responders. Key findings included:
- Optimal dispatch range: Expanding the dispatch range improves survival rates up to an optimal 800-meter range, beyond which the improvement is negligible.
- Impact of responder numbers: Increasing the number of responders from 20 to 100 significantly improves survival rates by up to 148%. However, beyond 100 responders, the improvement becomes marginal.
- Responder composition: Prioritizing skilled responders over mobile responders resulted in a 16% improvement in survival rates, while increasing mobile responders further improved outcomes by 3%.
Future research directions include further integration and optimization of GIS, considering the impact of traffic flow and special events on emergency response times, and dynamically adjusting deployment strategies.