ABSTRACT
Road crashes are events which depend on a variety of factors and which exhibit different magnitudes of outputs when evaluated with respect to the effects on road users. Despite a lot of research into the evaluation of crash likelihood and frequency, only a few works have focused exclusively on crash severity with these limited to sections of freeways and multilane highways. Hence, at present there is a large gap in knowledge on factors affecting the severity of crashes for other road categories, facilities, and scenarios. The paper deals with the identification of factors affecting crash severity level at urban road intersections. Two official crash records together with a weather database, a traffic data source with data aggregated into 5min intervals, and further information characterising the investigated urban intersections were used. Analyses were performed by using a back propagation neural network model and a generalized linear mixed model that enable the impact assessment of flow and other variables. Both methods demonstrate that flows play a role in the prediction of severity levels.
Subject(s)
Accidents, Traffic/statistics & numerical data , Factor Analysis, Statistical , Accidents, Traffic/mortality , Adolescent , Adult , Age Distribution , Age Factors , Databases, Factual , Environment Design , Female , Humans , Injury Severity Score , Linear Models , Male , Middle Aged , Reproducibility of Results , Sex Distribution , Weather , Young AdultABSTRACT
Traditional studies on road accidents estimate the effect of variables (such as vehicular flows, road geometry, vehicular characteristics), and the calculation of the number of accidents. A descriptive statistical analysis of the accidents (those used in the model) over the period 1992-1995 is proposed. The paper describes an alternative method based on the use of artificial neural networks (ANN) in order to work out a model that relates to the analysis of vehicular accidents in Milan. The degree of danger of urban intersections using different scenarios is quantified by the ANN model. Methodology is the first result, which allows us to tackle the modelling of urban vehicular accidents by the innovative use of ANN. Other results deal with model outputs: intersection complexity may determine a higher accident index depending on the regulation of intersection. The highest index for running over of pedestrian occurs at non-signalised intersections at night-time.
