Developing a Cluster-Based Algorithm for Collision Hotspot Identification
Traffic safety is one of the most important goals for roadway design and traffic system operations. Collision hotspot identification serves as a major fundamental component for traffic safety performance measurements. With identified collision hotspots, limited resources can be better allocated to improve roadway safety. There has been a significant amount of studies on collision hotspots identification over the past decades. However, most studies only considered crash counts as the sole roadway safety performance indicator. In this study, a new cluster-based method is proposed to quantify roadway safety conditions. This method is able to incorporate more heterogeneous safety-related factors for clustering, such as crash fatality, injuries, and average collision duration. Compared with the prevailing Empirical Bayes methods, the authors cluster-based method demonstrates its improved accuracy and efficiency, and can be easily implemented in practice.
- Record URL:
-
Availability:
- Find a library where document is available. Order URL: http://worldcat.org/isbn/9780784413623
-
Supplemental Notes:
- © 2014 American Society of Civil Engineers.
-
Corporate Authors:
American Society of Civil Engineers
1801 Alexander Bell Drive
Reston, VA United States 20191-4400 -
Authors:
- Bi, Chaofan
- Ma, Xiaolei
- Zhang, Yingying
- Wang, Yinhai
-
Conference:
- 14th COTA International Conference of Transportation Professionals
- Location: Changsha , China
- Date: 2014-7-4 to 2014-7-7
- Publication Date: 2014-6
Language
- English
Media Info
- Media Type: Web
- Features: References;
- Pagination: pp 2381-2395
- Monograph Title: CICTP 2014: Safe, Smart, and Sustainable Multimodal Transportation Systems
Subject/Index Terms
- TRT Terms: Crash data; Fatalities; High risk locations; Highway design; Injuries; Performance measurement; Traffic safety
- Subject Areas: Highways; Safety and Human Factors; I82: Accidents and Transport Infrastructure;
Filing Info
- Accession Number: 01531194
- Record Type: Publication
- ISBN: 9780784413623
- Files: TRIS, ASCE
- Created Date: Jul 24 2014 3:21PM