Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Considering the Reliability of Police-Reported Weather Information on Freeways Traffic Crash Severity Analysis: Proposing a Mixed Statistical and Geospatial Solution
Weather condition is one of the factors that influence the severity of traffic crashes. While, researchers in developing countries, have expressed concern about the quality and accuracy of police-reported information which includes weather status variable. This study presents a geospatial approach to spatially and temporally match weather data with police reports and use Mann–Whitney U, Kruskal–Wallis and two-way analysis of variance to identify effective factors on crash severity. Moreover, current research attempts to compare the effects of police-reported weather information (Scenario 1) with Meteorological Organization statistics (Scenario 2) on crash severity analysis. In this regard, the spatial kernel density estimation (KDE) method is applied on the crash records to evaluate crash prone sections according to the crash severity index (SI), as well as different weather conditions and effective temporal factors obtained from the results of the statistical analysis. The outputs of the two proposed scenarios showed different results in terms of the contribution of weather factors affecting property damage only crashes. The results revealed that the temporal factors such as holidays, together with extreme temperature conditions, has a significant impact on the severity of crashes. The results of spatial analysis of crash hotspots based on crash severity index displayed that the distribution of more severe crashes has increased on sections leading to mountainous regions, in snowy weather and in hot temperature conditions. The findings of this study indicated that the use of data-driven statistical and cluster-based spatial data mining techniques for analysis of large amounts of weather data in combination with police crash reports could lead to some new achievements on freeways crash severity analysis.
Considering the Reliability of Police-Reported Weather Information on Freeways Traffic Crash Severity Analysis: Proposing a Mixed Statistical and Geospatial Solution
Weather condition is one of the factors that influence the severity of traffic crashes. While, researchers in developing countries, have expressed concern about the quality and accuracy of police-reported information which includes weather status variable. This study presents a geospatial approach to spatially and temporally match weather data with police reports and use Mann–Whitney U, Kruskal–Wallis and two-way analysis of variance to identify effective factors on crash severity. Moreover, current research attempts to compare the effects of police-reported weather information (Scenario 1) with Meteorological Organization statistics (Scenario 2) on crash severity analysis. In this regard, the spatial kernel density estimation (KDE) method is applied on the crash records to evaluate crash prone sections according to the crash severity index (SI), as well as different weather conditions and effective temporal factors obtained from the results of the statistical analysis. The outputs of the two proposed scenarios showed different results in terms of the contribution of weather factors affecting property damage only crashes. The results revealed that the temporal factors such as holidays, together with extreme temperature conditions, has a significant impact on the severity of crashes. The results of spatial analysis of crash hotspots based on crash severity index displayed that the distribution of more severe crashes has increased on sections leading to mountainous regions, in snowy weather and in hot temperature conditions. The findings of this study indicated that the use of data-driven statistical and cluster-based spatial data mining techniques for analysis of large amounts of weather data in combination with police crash reports could lead to some new achievements on freeways crash severity analysis.
Considering the Reliability of Police-Reported Weather Information on Freeways Traffic Crash Severity Analysis: Proposing a Mixed Statistical and Geospatial Solution
Transp. in Dev. Econ.
Effati, Meysam (Autor:in) / Atrchian, Chakavak (Autor:in)
01.10.2024
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Analysis of Crash Precursors on Instrumented Freeways
| British Library Online Contents | 2002
Novel Crash Surrogate Measure for Freeways
| ASCE | 2020
Accuracy of Injury Severity Ratings on Police Crash Reports
| British Library Online Contents | 2015
Engineering Index Backfile | 1960