What Role Do Precrash Driver Actions Play in Work Zone Crashes? Application of Hierarchical Models to Crash Data

Highway infrastructure requires periodic maintenance, reconstruction, and rehabilitation. As a result, highway users have to deal with work zone activities such as lane closures and lane shifts or crossovers, and they must be aware of changes in road conditions to drive safely through work zones. A large-scale statewide crash database from the Virginia Department of Transportation was used in this study to examine correlations between precrash actions and the severity of the injuries of drivers involved in crashes. An innovative aspect of this study is that it accounts for the injury severity of each vehicle driver involved in a crash by estimating hierarchical models. The correlates of the severity of a driver’s injury are nested in crashes. Modeling revealed that for work zone crashes the chances of driver injury are 9.9% to 10.3% higher than for the base behavior (i.e., no improper actions by the driver) if the driver intentionally commits an improper action or a violation. For non–work zone crashes, the chances of injury are higher by only 1.7% to 5.7% compared with the base behavior. Such actions and violations are mainly the following: (a) speeding, (b) following too closely, and (c) disregarding officers, flaggers, signals, and signs. The correlations between precrash actions and injury severity provide insights into safety improvements (e.g., effective speed enforcement and traffic regulations) that could reduce the risk of injury in work zones. Hierarchies embedded in highway crash data were explored in this study to make methodological and empirical contributions to the understanding of work zone safety and to improve understanding of behaviors that lead to injuries and fatalities in work zone crashes.


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01595009
  • Record Type: Publication
  • ISBN: 9780309441193
  • Files: TRIS, TRB, ATRI
  • Created Date: Mar 30 2016 2:20PM