1. Home
  2. TRID
  3. View Record
https://www.nationalacademies.org/trb/events

Reducing Crash Risk at Work Zones in South Carolina

This project investigated several aspects related to work zone safety. First, it examined the accuracy of the information recorded in the South Carolina (SC) traffic collision report forms (TR-310). A total of 200 fatal crashes in work zones between 2014 and 2020 were examined to determine how many discrepancies exist between the written narrative and other fields in the traffic collision forms. Of the 200 forms, 63.5%, 31%, and 5.5% contained 0, 1, and 2 discrepancies, respectively. Second, it determined factors that contributed to injury in work zone-related crashes in SC on roads with 60 mph or higher speed limits and those with speed limits less than 60 mph using crash data from 2014 to 2020 in SC. Results from mixed binary (injury or no injury) logit models indicate there are common factors: dark lighting conditions, female (at-fault) drivers, and excessive speed. Factors that contributed to injury on roads with speed limits less than 60 mph are work zone on an SC or US primary road, work zone activity area, at-fault drivers under 35, sideswipe collisions, and worker presence. Contributing factors on roads with 60 mph or higher speed limits are number of vehicles involved, rear-end collisions, proximity to first work zone signs, and weekdays. Additionally, a mixed binary (injury or no injury) logit model with heterogeneity in mean and variance was used to determine factors that contributed to injury severity of work zone rear-end crashes with collision speeds over 35 mph. Factors that increased injury severity include multi-vehicle involvement, airbag deployment, dark conditions, and truck involvement. Conversely, factors that reduced injury severity are late-night and dawn/dusk conditions, advanced warning areas, work zone activity areas, lane shifts/crossover work zones, and young and middle-aged at-fault drivers. Third, it assessed the effectiveness of the presence of law enforcement at work zones. Using a split-plot design with blocking, eight models were explored. These models use two types of response variables, average speed in the entire work zone and average speed in the transition area. For each type of response, two variations were assessed. The first is to subtract the temporary posted speed limit from the average speed, and the other variation is to include traffic volume as a covariate. The results indicated that traffic speeds at work zones were lower when law enforcement was present. Fourth and lastly, it developed a predictive work zone risk assessment tool to enable South Carolina Department of Transportation (SCDOT) engineers to determine crash risk and benefit-cost of implementing countermeasures. This tool was implemented in Excel using VBA and estimation results from a zero-inflated negative binomial model to predict the expected number of crashes given a work zone’s length, duration, and Annual Average Daily Traffic (AADT).

  • Record URL:
  • Corporate Authors:

    University of South Carolina, Columbia

    Department of Civil and Environmental Engineering, 300 South Main Street
    Columbia, SC  United States  29208

    South Carolina Department of Transportation

    Columbia, SC  United States  29202

    Federal Highway Administration

    1200 New Jersey Avenue, SE
    Washington, DC  United States  20590
  • Authors:
    • Huynh, Nathan
    • Madarshahian, Mahyar
    • Wegmet, Jackson
  • Publication Date: 2024-6

Language

  • English

Media Info

  • Media Type: Digital/other
  • Edition: Final Report
  • Features: Appendices; Figures; References; Tables;
  • Pagination: 84p

Subject/Index Terms

Filing Info

  • Accession Number: 01933115
  • Record Type: Publication
  • Report/Paper Numbers: FHWA-SC-24-04
  • Contract Numbers: SPR No. 760
  • Files: NTL, TRIS, USDOT, STATEDOT
  • Created Date: Oct 9 2024 4:01PM