Identification and priorization techniques aplied to urban traffic crash locations: A scoping review

Abstract

This paper presents a Scoping Review of methods used for the identification and prioritization of Traffic Crash Locations (TCL) in complex urban areas, considering different accident classes and outcomes with different degrees of severity. Were obtained relevant studies on identifying and prioritizing accident-prone locations from the consultation of two electronic databases (Scopus and Web of Science). The data obtained were evaluated by independent reviewers. Studies carried out in urban areas, which carry out the tasks of grouping accidents into units of analysis, identifying candidate areas, and prioritizing accident-prone locations, were selected. Forty-two studies were selected and evaluated. The applied units of analysis, identification methods, and criteria for prioritizing candidate areas for accident-prone locations were identified among the studies found. The mapping of the works that apply statistical significance criteria of the findings and longitudinal analyses of the candidate sites was carried out. Among the strategies used to identify TCL, the most frequently applied approach was Spatial Association (SA) (23.81%), followed by minimum threshold limit (MTL) (14.29%). Among the different ranking criteria identified, the upper control limit (UCL) criterion stood out, followed by the density and expected frequencies criteria. This work presents a narrative summary mapping the heterogeneity of applied approaches. The contribution of this study was the presentation of the roadmap for the different studies related to the analysis of TCL. Finally, this work points out the need to adopt longitudinal analyzes as a criterion for prioritizing TCL and the application of criteria of statistical significance during the prioritization stage of TCL.