UTC 2022 Funding - Cycle 1 Research Projects

Project No.: CY1-OSU-02
Title: Fast Detection and Prediction of Slippery Roadway Conditions for Enhanced Safety
Performing Institution: Oklahoma State University
Principal Investigator: Joshua Q. Li
Start and Anticipated Completion Dates: 10/1/2023-9/30/2024
Abstract: Black ice, a nearly invisible hazard, contributes to over 10% of weather-related crashes in the U.S., causing 200,000 annual accidents, 700 fatalities, and 65,000 injuries. Traditional methods for detecting black ice involve fixed sensors and signs, but new vehicle-based technology offers cost-effective real-time data. However, obtaining comprehensive road condition data during inclement weather remains expensive and risky. State agencies must collect pavement surface data for asset management, yet the relationships between surface characteristics, weather conditions, and ice formation are not adequately understood. Research is needed to predict slippery conditions using existing data. Prediction of slippery conditions can be potentially more critical than detecting slippery conditions due to changing climates and weather extremes.

This project aims to develop predictive models for slippery road conditions by collecting data with Mobile Advanced Road Weather Information Sensors (MARWIS) sensors and Pave3D 8K on roadway segments before, during, and after inclement weather. The collected data will be used to create predictive models for different weather scenarios. The primary goal is to develop predictive models that can anticipate slippery road conditions under different weather scenarios. These prediction models can then be applied to identify potentially slippery areas across Oklahoma, using the annually collected PMS datasets by ODOT. The primary goal of this project is to enhance highway safety.

The aforementioned goals will be achieved through four tasks: Task 1: Data Collection: Use MARWIS technology to measure road conditions, including temperature, humidity, and road state. This data will be collected on selected testing sites based on weather forecasts and in collaboration with ODOT; Task 2: Surface Characteristics: Assess field friction values and collect pavement surface characteristics data using the Grip Tester and Pave3D 8K technology to understand their impact on road slipperiness; Task 3: Slippery Road Prediction Models: Leverage data from MARWIS and surface characteristics and create predictive models using statistical and machine learning methods for forecasting road conditions during rainy or icy days; Task 4: Implementation: Incorporate statewide surface characteristics data from ODOT into the predictive models, presenting results in a Geographic Information System (GIS) database for better situational awareness and road maintenance support.
Click to read more