A case study on the application of an approximated hypercube model to emergency medical systems management

This paper describes an application of the approximated hypercube model to Lisbon emergency medical services (EMS) management, namely for assessing alternative dispatching rules for assigning ambulances to emergency calls. The approximated hypercube (A-hypercube) is a queuing theory model that computes several performance measures such as average response time, server workloads or the probability of all servers being busy (loss probability). The assumptions of the extended model are Poisson customer arrivals, general service time (customer and server dependent) and a fixed preference assignment rule of servers to customers. The fact that dispatching rules are precisely a model parameter, turn this model into a valuable tool in the definition of efficient operating rules. In this paper, we propose new expressions for the computation of system performance measures during periods in which the emergency call arrival process is not stationary. Different dispatching rules are evaluated by comparing the system performance measures obtained from the extended A-hypercube model and a simulation model, using data collected from the Lisbon EMS Department.