|
Activity Number:
|
353
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Social Statistics Section
|
| Abstract - #301933 |
|
Title:
|
Safety Analyses of Signalized Intersections Using Bayesian Hierarchical Spatial Models
|
|
Author(s):
|
Feng Guo*+ and Xuesong Wang and Mohamed Abdel-Aty
|
|
Companies:
|
Virginia Polytechnic Institute and State University and University of Central Florida and University of Centraol Florida
|
|
Address:
|
415 B Hutcheson Hall, Blacksburg, VA, 24061,
|
|
Keywords:
|
Bayesian ; Spatial ; Signalized intersection ; CAR ; Traffic Safety
|
|
Abstract:
|
Signalized intersections are among the most dangerous locations on a highway network. There are more than 2.8 million intersection-related crashes in year 2000. In this paper we propose a Bayesian hierarchical approach to model accidents at signalized intersections. The data include 197 signalized intersections in the State of Florida. The proposed model incorporates the effects of geometric design features, traffic control features, and traffic characteristics. Furthermore, we argue that signalized intersections along a certain corridor will affect each other, especially for those in close proximity. We incorporate this spatial correlation into the model by using a conditional autoregressive prior for an intersection specific random effect. The results of this study will benefit intersection geometric design and traffic management.
|
