An Intelligent Feedback Approach for Locating and Dispatching Two Types of Ambulances under Uncertainty and Partial Coverage Considerations

Abstract: The Emergency Vehicle Covering and Planning (EVCP) problem locates a limited number of two heterogeneous types of ambulances in different city points and dispatches them to the emergency scenes, considering the uncertainty of the emergency locations, to maximize the emergency total and partial coverage and the response time in which the patients receive medical first aids. We propose a novel two-stage quadratic stochastic program for the EVCP problem that locates the limited number of heterogeneous types of ambulances in the first stage. The second stage deals with dispatching ambulances to accidents. The EVCP stochastic model allows partial coverage of the accidents by the ambulances based on a decay function. Instead of decomposing the stochastic model, we propose a location-allocation methodology that relies on the solution of an auxiliary surrogate model, which is faster to solve. The location of the ambulances obtained by this surrogate model is input to the original model. Experimental results show that we obtain high-quality solutions in a reasonable time.