Fast Detection and Prediction of Slippery Roadway Conditions for Enhanced Safety

Details

• Creators:
  Li, Joshua Qiang ; Munna, Masud Rana
• Corporate Creators:
  Oklahoma State University
• Corporate Contributors:
  The Southern Plains Transportation Center (SPTC) Region 6 University Transportation Center (UTC) ; United States. Department of Transportation. Office of the Assistant Secretary for Research and Technology
• Subject/TRT Terms:
  Black Ice Detection And Identification Predictive Models Road Weather Information Systems Slipperiness Weather Conditions
• Publication/ Report Number:
  CY1-OSU-02
• Resource Type:
  Tech Report
• Geographical Coverage:
  United States
• Edition:
  Final Report (Dec 2023- Dec 2024)
• Corporate Publisher:
  The Southern Plains Transportation Center (SPTC) Region 6 University Transportation Center (UTC)
• Abstract:
  Black ice, a nearly invisible hazard causing over 10% of weather-related crashes in the US, poses significant risks to roadway safety. Current measures like stationary Road Weather Information sensors (mRWIS) and cautionary signs improve awareness, but have limitations. Mobile Advanced Road Weather Information Sensors (MARWIS) offer enhanced, cost-effective, and real-time data collection at highway speed at the network level during and after inclement weather. The project aimed to develop procedures and models for fast detection and prediction of slippery conditions using the data of 51 miles of roadway, including SH-51, SH-177, SH-33, and a county road in Oklahoma. In conjunction with surface characteristics, the project was divided into three major steps on roadway conditions data collection by using MARWIS as measured parameters like ice percentage, water film height, road temperature, humidity, and pavement condition. In the middle, using Pave 3D 8K technology the project captured detailed pavement surface characteristics, and pavement friction by using a Grip Tester. Finally, machine learning methods were used (i.e., regression model, Random Forest, and Gradient Boosting) to analyze the data by modeling ice percentage, water film height, and pavement friction. The models achieved accuracy rates up to 75.8% and highlighted critical factors influencing icy and rainy conditions using the key variables of Mean Profile Depth (MPD), International Roughness Index (IRI), Cross Slope, and Crack Density. These findings enhance proactive safety measures and align with USDOT priorities for road safety.
• Format:
  PDF
• Funding:
  69A3552348306
• Collection(s):
  University Transportation Centers
• Main Document Checksum:
  urn:sha-512:f5f4518ff5bd9acbb0346ff0357461aa7dc11a573679b62b6ff014bac400b5d3188094994061dd3c9f58a830a4ae5d15a60b05cde2b3408614ba57ec2b781838
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/88402/dot_88402_DS1.pdf
• File Type:
  [PDF - 8.91 MB ]