Vehicular Flow Analysis Using Clusters

Abstract

The volume of vehicular traffic in large cities has increased in recent years, causing mobility problems, which is why the analysis of vehicular flow data becomes important for researchers. Intelligent Transportation Systems perform vehicle monitoring and control by collecting GPS trajectories, information that provides real-time geographic location of vehicles, which allows the identification of patterns on vehicle flow using clustering techniques. This paper presents a methodology capable of analyzing vehicular flow in a given area, identifying speed ranges and maintaining an updated interactive map that facilitates the identification of areas of possible traffic jams. The results obtained on a dataset from the city of Guayaquil-Ecuador are satisfactory and clearly represent the speed of vehicle displacement by automatically identifying the most representative ranges for each instant of time.