Traffic accident is one the most substantial concern of road safety. Prediction models are being used for defining relations between freeway accidents and effective parameters such as traffic volume, road geometric designs and environmental issues. In this research work, artificial neural network and log-normal models have been proposed to estimate the number of road accidents in Tehran-Qom freeway. Average daily traffic volume, percentage of heavy vehicle, average speed and pavement condition index are considered as input variables. Three-year period of accident data and parameters, have been used in analytical process modeling and validation. Efficiency ranking of variables in artificial neural networks and log-normal regression extracted based on the coefficients of determination. Results show that average speed of vehicles and average daily traffic volume are the most effective parameters in freeway accidents as well as artificial neural network model is more capable to estimate the number of road accidents in freeways.
Variable Efficiency Appraisal in Freeway Accidents Using Artificial Neural Networks—Case Study
The Twelfth COTA International Conference of Transportation Professionals ; 2012 ; Beijing, China
CICTP 2012 ; 2657-2664
2012-07-23
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Freeway ramp metering using artificial neural networks
Online Contents | 1997
|Artificial Neural Networks for Freeway Incident Detection
Online Contents | 1995
|Congestion management of freeway traffic using artificial neural networks
British Library Conference Proceedings | 1997
|Modelling freeway interchange accidents
Elsevier | 1991
NTIS | 1990
|