Please use this identifier to cite or link to this item:
Title: Applying Bayesian hierarchical models to examine motorcycle crashes at signalized intersections
Authors: Haque, M.M. 
Chin, H.C. 
Huang, H.
Keywords: Bayesian inference
Four-legged intersections
Hierarchical models
Motorcycle crashes
T intersections
Issue Date: Jan-2010
Citation: Haque, M.M., Chin, H.C., Huang, H. (2010-01). Applying Bayesian hierarchical models to examine motorcycle crashes at signalized intersections. Accident Analysis and Prevention 42 (1) : 203-212. ScholarBank@NUS Repository.
Abstract: Motorcycles are overrepresented in road traffic crashes and particularly vulnerable at signalized intersections. The objective of this study is to identify causal factors affecting the motorcycle crashes at both four-legged and T signalized intersections. Treating the data in time-series cross-section panels, this study explores different Hierarchical Poisson models and found that the model allowing autoregressive lag-1 dependence specification in the error term is the most suitable. Results show that the number of lanes at the four-legged signalized intersections significantly increases motorcycle crashes largely because of the higher exposure resulting from higher motorcycle accumulation at the stop line. Furthermore, the presence of a wide median and an uncontrolled left-turn lane at major roadways of four-legged intersections exacerbate this potential hazard. For T signalized intersections, the presence of exclusive right-turn lane at both major and minor roadways and an uncontrolled left-turn lane at major roadways increases motorcycle crashes. Motorcycle crashes increase on high-speed roadways because they are more vulnerable and less likely to react in time during conflicts. The presence of red light cameras reduces motorcycle crashes significantly for both four-legged and T intersections. With the red light camera, motorcycles are less exposed to conflicts because it is observed that they are more disciplined in queuing at the stop line and less likely to jump start at the start of green. © 2009 Elsevier Ltd. All rights reserved.
Source Title: Accident Analysis and Prevention
ISSN: 00014575
DOI: 10.1016/j.aap.2009.07.022
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.


checked on Oct 26, 2021


checked on Oct 26, 2021

Page view(s)

checked on Oct 28, 2021

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.