Search the digital library catalog

Abstract: Despite an improvement in worldwide numbers, road traffic crashes still cause social, psychological, and financial damage and cost most countries 3% of their gross domestic product. However, none of the current commercial or open-source navigation systems contain spatial information about road traffic crash hot spots. By developing an algorithm that can adequately predict such spatial patterns, we can bridge these still existing gaps in road traffic safety. To that end, geographically weighted regression and regression tree models were fitted with five uncorrelated (environmental and socioeconomic) road traffic crash predictor variables. Significant regional differences in adverse weather conditions were identified; Slovenia lies at the conjunction of different climatic zones characterized by differences in weather phenomena, which further modify traffic safety. Thus, more attention to speed limits, safety distance, and other vehicles entering and leaving the system could be expected. In order to further improve road safety and better implement globally sustainable development goals, studies with applicative solutions are urgently needed. Modern vehicle-to-vehicle communication technologies could soon support drivers with real-time traffic data and thus potentially prevent road network crashes.
Keywords: GIS, hot spot analyses, traffic safety, spatial modelling, weather patterns
Published in DKUM: 20.08.2024; Views: 92; Downloads: 9
Full text (1,42 MB)
This document has many files! More...