INFLUENCE OF DIFFERENT DISTANCE METRICS, ON EMERGENCY VEHICLES RELOCATION OPTIMIZATION PROCESS, BASED ON AUTOMATIC VEHICLE LOCATION DATA

Abstract

In the implementation of the emergency vehicle dwelling location optimization process, different approaches in definition of the distances between two locations are possible. The distances can be treated as Euclidian, road network routed or time distances. The main question is how the choice of different type of distances used affects  the quality of the optimization solution. In this study, using the same data set and identical method, three different types of distances are used and the results are compared. For the purpose of comparison and evaluation of the quality of the obtained solutions, an innovative definition of a parameter named the saving ratio is proposed. On the basis of that parameter, quality of the obtained optimization solutions is quantified and estimated. In the previous works of the same authors, extensive algorithm to resolve the problem of p-median is designed. That algorithm, named GA-GISLAB, is used as a test bed optimization method. It performs relocation of vehicle resources in the road traffic based on the modified Genetic algorithm. Assessment of the impact of different types of distances to the quality of the optimization solution was performed using historical data from automatic vehicle location system.