Citation

Wu J. Transp. Res. Circular 2017; (E-C221): 24-38.

Copyright

(Copyright © 2017, U.S. National Academy of Sciences Transportation Research Board)

DOI

unavailable

PMID

unavailable

Abstract

Roadway crashes have been a major concern for public safety. One of the main issues in traffic safety is the conflict between turning vehicles and other roadway users at intersections. Turning driver behavior has not been studied in depth due to limited data. In this paper, driver behavior was provided based on the second Strategic Highway Research Program Naturalistic Driving Study. A total of 211 right-turn crash and near-crash turning events at signalized intersections were used to analyze driver behavior. Each turning movement was divided into different stages by analyzing the sensor data and by reviewing the forward videos. Driver behavior was extracted from crash detailed information provided by the database. Prioritized influencing factors on crash severity were identified by using the random forest method. Driver behavior and speed were the top two significant influencing factors on crash severity. An ordered probit model was applied to the risk factors on turning crash severity at intersections. The results showed distraction behavior, high speed, and deceleration were associated with the most serious right-turn crashes at intersections. The relationship between driver behavior and traffic flow, roadway geometric design, and traffic signs were also analyzed in this paper. The result showed that in a "right turn on red" situation, 80% of drivers involved in the right-turn events did not fully stop before turning right. Driving operation varied with different geometric designs of intersections. Drivers were found to have the lowest frequency of improper operation when conflicting traffic flow was between 1,000 vph and 2,000 vph. As face videos and baseline data were not available for this research, more related data is still needed to evaluate the findings.

Language: en

Keywords

Behavior; Drivers; Signalized intersections; Crash severity; Automatic data collection systems; Geometric design; Right turns; Human factors in crashes; Strategic Highway Research Program 2