Optimal location for a helicopter in a rural trauma system: prediction using discrete-event computer simulation.

A discrete-event computer simulation was developed using the C programming language to determine the optimal base location for a trauma system helicopter in Maine, a rural area with unevenly distributed population. Ambulance run reports from a one-year period provided input data on the times and places where major injuries occurred. Data from a statewide trauma registry were used to estimate the percentage of cases which would require trauma center care and the locations of functional trauma centers. Climatic data for this region were used to estimate the likelihood that a helicopter could not fly due to bad weather. The incidence of trauma events was modeled as a nonstationary Poisson process, and location of the events by an empirical distribution. For each simulated event, if the injuries were sufficiently severe, if weather permitted flying, if the occurrence were not within 20 miles of a center or outside the range of the helicopter, and if the helicopter were not already in service, then it was used for transportation. 35 simulated years were run for each of 4 proposed locations for the helicopter base. One of the geographically intermediate locations was shown to produce the most frequent utilization of the helicopter. Discrete-event simulation is a potentially useful tool in planning for emergency medical services systems. Further refinements and validation of predictions may lead to wider utilization.