TY  - JOUR
PY  - 2006//
TI  - Stochastic modelling for traffic crashes on non urban highways in India
JO  - Proceedings: ARRB Group Biennial Conference
A1  - Parida, M
A1  - Jain, SS
A1  - Landge, VS
SP  - 
EP  - 
VL  - 22
IS  - CD-ROM
N2  - In this paper an attempt has been made to develop correlation between highway geometric parameters, traffic parameters and  crash rate for two lane non urban highways.  To simulate the  stochastic nature of crash occurrence, Poisson regression and  negative binomial regression has been employed.  A detailed  study has been carried out on a two lane non urban highway  (National Highway No 58) in India.  The study included  collecting of traffic and geometric features of the highway along  with past crash data for 8 years.  Variables included in the  model are access density per km, curve density per km,  intersection density per km, space mean speed, shoulder width,  volume of heavy vehicles, volume of non-motorized and animal  drawn vehicles and total volume.  The dependent variable is  crash rate per year per km.  Strong positive relationships were  observed between volume of heavy vehicle and crashes and  speed and crash rate.  Shoulder width and lane width show  negative relationship with crash rate.
LA  - 
SN  - 0572-1431
UR  - http://dx.doi.org/
ID  - ref1
ER  -