An iterative method for simultaneously locating trauma centers and helicopters through the planning horizon

Abstract

This study addresses the problem of simultaneously locating trauma centers and helicopters during the planning horizon. Motivated by a prior study (Cho et al., 2014), we extend the location model into a multi-period location model by introducing an additional decision point: when to locate. Because the original location model is known to be quite difficult given its unique characteristic, in which trauma centers and helicopter locations are coupled, a location model with time-dependent parameters becomes even more difficult to solve. Specifically, the complicated nature of the problem arises from estimating the availability of helicopters, which is required to solve the location model. To tackle this difficulty, we propose an efficient solution approach that iteratively updates helicopters availability using the previous step of optimization results. At each iteration, the optimization results, including location solutions, pass to the stage of calculating each helicopter's busy fraction, and are fed into the parameters of the location model for the next round of optimization until the termination condition is met. Using a series of experimental results with real data gathered from Korean trauma cases, we confirm that the proposed method is valid and competitive compared to alternative methods. (C) 2018 Elsevier Ltd. All rights reserved.