Cost-effective analyses for emergency medical services deployment: A case study in Shanghai

The response speed of emergency medical services (EMS) is extremely critical for pre-hospital lifesaving. Each second plays a vital role in emergent cases. Therefore, the EMS network should be carefully designed to provide timely and reliable services. This paper focuses on improving the classical ambulance deployment method to optimize the distribution of first-aid resources. We refine the double coverage model with a new objective function that minimizes the expected cost for delayed emergencies plus the operational cost for EMS. EMS operators can use this refined model to analyze the cost-effectiveness of the emergency network, as demonstrated in a real-world EMS deployment case study. To solve this large-scale integer-programming problem, we compared several heuristic algorithms and finally chose ant colony optimization algorithm. Several modifications are made to adapt the traditional ant colony algorithm to solve our problem. The paper concludes with an optimized and cost-effective plan for ambulance deployment in the downtown area of Shanghai. After optimization, the objective cost is lowered by about 11% each year and 10% less emergency calls are missed or delayed. In addition, we use our modified model to test scenarios with relaxed limit of resources. Potential improvement of future EMS network is discovered. It turns out that increasing the number of first-aid stations from 35 (current number of stations) to 50 and adding 10 vehicles will achieve a surprising 49% cost saving while maintaining a satisfactory service level.