An accident prediction model for urban road networks

For several years now increasingly analytical prediction models have been developed that are able to correlate accident frequency with infrastructure characteristics to support the planning and design of measures for enhancing road safety. The models developed so far, though useful in the context fo...

Full description

Saved in:
Bibliographic Details
Published in:Journal of transportation safety & security 2018-07, Vol.10 (4), p.387-405
Main Authors: Gianfranco, Fancello, Soddu, Stefano, Fadda, Paolo
Format: Article
Language:English
Subjects:
accident prediction
Accidents
field research
Poisson distribution
Prediction models
Road construction
Roads
Roads & highways
Traffic accidents & safety
urban accidents
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:For several years now increasingly analytical prediction models have been developed that are able to correlate accident frequency with infrastructure characteristics to support the planning and design of measures for enhancing road safety. The models developed so far, though useful in the context for which they have been calibrated, are limited by the fact that they are not transferable to other contexts because of different traffic regulations. The present work aims to develop a predictive model for urban roads that is able to estimate the number of accidents for three situations in an urban road network, a roundabout, a three- or four-way junction, and a straight stretch of road. The models constructed are based on Poisson's and negative binomial algorithms and can be readily applied for accident prediction or identification of black spots.
ISSN:1943-9962
1943-9970
DOI:10.1080/19439962.2016.1268659