CLOSE OVERLAY

Assessing Driver Behavior at Back of Queues: Implications for Queue Warning System in Work Zones

Project Details
STATUS

Completed

PROJECT NUMBER

19-686, TPF-5(295)

START DATE

01/01/19

END DATE

07/24/20

FOCUS AREAS

Safety

RESEARCH CENTERS InTrans, CTRE, SWZDI
SPONSORS

Iowa Department of Transportation
Smart Work Zone Deployment Initiative

Researchers
Principal Investigator
Shauna Hallmark

Director, InTrans

Co-Principal Investigator
Anuj Sharma

Research Scientist and Leader, REACTOR

About the research

Rear-end crashes are one of the primary crash types in work zones and frequently occur at the back-of-queue (BOQ). Some agencies have utilized back-of-queue warning systems (QWSs), where real-time sensors are located upstream of stopped or slowed traffic, either to actually detect BOQs or monitor conditions to predict BOQ locations. QWSs then provide notifications of traffic conditions to drivers, which ideally lead to lower speeds and drivers being prepared to react to the BOQ, resulting in fewer crashes and conflicts. However, a driver needs to be properly monitoring the roadway environment to receive the warning and, then, needs to be prepared to take the appropriate actions when necessary. In many cases, drivers are distracted and fail to recognize warnings, or they receive the warning but fail to comply with appropriate speeds. As a result, one of the main needs to address BOQ situations is to understand what drivers are doing so that a QWS can get a driver’s attention. Additionally, driver behavior may indicate that other countermeasures, such as speed management, may be as effective as formal QWSs. The research described in this report aims to address this knowledge gap through the following objectives:

  • Identify common types of QWSs
  • Summarize QWSs used in Smart Work Zone Deployment Initiative (SWZDI) states
  • Identify driver behaviors in BOQ scenarios
  • Make recommendations
  • Summarize needs for connected vehicle applications

Safety critical events (SCEs) were evaluated for back-of-queue situations using two different datasets. The first was a set of BOQ SCEs that were reduced from camera image captures at BOQ locations in work zones in Iowa during the 2019 construction season. Analysis of these data indicated speeding, following too closely, and forced merges were the primary characteristics associated with BOQ. The second dataset was an analysis of BOQ events in the second Strategic Highway Research Program (SHRP2) Naturalistic Driving Study (NDS). Analysis of these data indicated that following too closely and glances away from the roadway task of 1 or more seconds were statistically significant.

TOP