All Accidents are Not Equal: Using Geographically Weighted Regressions Models to Assess and Forecast Accident Impacts
Transportation professionals have long recognized the importance of accounting for accident and incident impacts when designing and constructing transportation networks. Studies have used statistical models of accidents to explore associations of traffic injuries/harm with driver, vehicle, roadway and environmental factors. A critical characteristic of most studies is their reliance on traditional models (referred to as global models) that assume the efficacy of a single set of estimated parameters to forecast crash impacts — an approach characterized as “one size fits all.” However, in spatially diverse metropolitan regions, accident impacts and their associations with variables can vary across space, resulting in unobserved spatial heterogeneity. Overcoming this weakness has led to the use of various spatial analysis techniques. Using collision data from the Hampton Roads region of southeastern Virginia and the estimated economic costs of accidents, this paper explores spatial relationships and provides comparisons of results obtained from global models and those obtained using the technique of Geographically Weighted Regression (GWR). Estimation of collision harm (assumed to be related to the approximate monetary costs of accidents) indicates that GWR methods yield significantly more accurate results. The results provide valuable information on high-risk factors associated with collision harm and the spatial variations in these associations and suggest improved data application in dynamic traffic simulations.
- Record URL:
-
-
Supplemental Notes:
- Distribution, posting, or copying of this PDF is strictly prohibited without written permission of the Transportation Research Board of the National Academy of Sciences. Unless otherwise indicated, all materials in this PDF are copyrighted by the National Academy of Sciences. Copyright © National Academy of Sciences. All rights reserved
-
Corporate Authors:
500 Fifth Street, NW
Washington, DC United States 20001 -
Authors:
- Zheng, Libing
- Robinson, R Michael
- Khattak, Asad J
- Wang, Xin
-
Conference:
- 3rd International Conference on Road Safety and Simulation
- Location: Indianapolis Indiana, United States
- Date: 2011-9-14 to 2011-9-16
- Publication Date: 2011
Language
- English
Media Info
- Media Type: Digital/other
- Features: Maps; References; Tables;
- Pagination: 13p
- Monograph Title: 3rd International Conference on Road Safety and Simulation
Subject/Index Terms
- TRT Terms: Crash data; Crash injuries; High risk locations; Incident management; Traffic crashes; Traffic simulation
- Uncontrolled Terms: Geographically weighted regression
- Geographic Terms: Hampton Roads (Virginia)
- Subject Areas: Highways; Planning and Forecasting; Safety and Human Factors; I72: Traffic and Transport Planning; I80: Accident Studies;
Filing Info
- Accession Number: 01504439
- Record Type: Publication
- Files: TRIS, TRB, ATRI
- Created Date: Jan 24 2014 2:29PM