Spatial Analysis with Weighted Kernel Groupings to Determine Risk Sectors Due to Traffic Accidents in Urban Area. Tunja Analysis, Colombia

Abstract

A method is presented to identify and determine groups with risk sectors due to the greater occurrence of traffic accidents in urban areas as an integral component in road safety management. The methodology was framed in Spatial Analysis with geographic statistics based on Exploratory Data Analysis (AED), Kernel Density Estimation (KDE), and the application of correlation and geoprocessing techniques. The accident data collected between 2015 and 2018 from the urban area of ​​Tunja, Boyacá, Colombia, were the basis for the study of the distribution of events, characterization of clusters, occurrence dynamics and pattern modeling. The definition and delimitation of risks depended on the dispersion or grouping (Hotspots) found with weighted Kernel together with the socio-spatial interrelation of underlying processes due to the territorial dynamics of the sector. The results reveal patterns of events in concentration foci with different levels of risk, in which land uses of opposite characteristics coexist according to their activities [commercial and residential], socioeconomic sectors of low strata with a mixture of arterial road network that by its functionality mobilizes high vehicular and pedestrian flows. Although the analysis is limited to a case study, the findings show a promising perspective in road safety by delimiting risk sites for traffic accidents through the incidence of territorial variables.