Proactive Identification of Hazardous Traffic Conditions Caused by Reduced Visibility Using Road Weather Information
This study proposes a novel methodology to identify hazardous freeway traffic conditions based on the prediction of real-time weather and traffic data obtained from road weather information systems (RWIS) and vehicle detection systems (VDS), respectively. The proposed methodology consists of two components. The first one is to develop a novel algorithm to derive an index representing hazardous traffic conditions by comparing visibility distance with safe stopping distance. The other component is to produce predictive information to represent the level of traffic stream safety based on the proposed index. A K-nearest neighbor method using historical patterns was adopted to obtain predictive visibility distance and safe stopping distance. The resultant cross validations showed that the mean absolute percent error is 7.43 percent on average, which demonstrated the prediction capability of the K-nearest neighbors method. The outcome of this study provides valuable weather-related traffic information so that traffic operations agents can actively manage traffic conditions for traffic safety enhancement. This study can also be used to support advanced warning or monitoring system implementation to avoid crashes under limited visibility conditions.
-
Supplemental Notes:
- This paper was sponsored by TRB committee AH010 Standing Committee on Surface Transportation Weather.
-
Corporate Authors:
500 Fifth Street, NW
Washington, DC United States 20001 -
Authors:
- Park, Hyunjin
- Jung, Soyoung
- Oh, Cheol
-
Conference:
- Transportation Research Board 96th Annual Meeting
- Location: Washington DC, United States
- Date: 2017-1-8 to 2017-1-12
- Date: 2017
Language
- English
Media Info
- Media Type: Digital/other
- Features: Figures; References; Tables;
- Pagination: 18p
- Monograph Title: TRB 96th Annual Meeting Compendium of Papers
Subject/Index Terms
- TRT Terms: Algorithms; Automatic vehicle detection and identification systems; Methodology; Pattern recognition systems; Road weather information systems; Stopping distances; Visibility distance
- Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management;
Filing Info
- Accession Number: 01622847
- Record Type: Publication
- Report/Paper Numbers: 17-01062
- Files: TRIS, TRB, ATRI
- Created Date: Jan 24 2017 9:07AM