Effects of Pavement and Weather Conditions on Roadway Safety

Abstract: Abstract: Advancements in sensor technology now enable real-time acquisition of comprehensive road conditions and weather data across extensive networks during and following inclement weather events. This study specifically aims to analyze how pavement conditions, roadway geometric factors, and various weather parameters influence roadway safety under adverse conditions. Field data were gathered at selected testing locations using a Mobile Advanced Road Weather Information System (MARWIS) for real-time climate data collection, cutting-edge 3D laser imaging for detailed pavement surface condition analysis, the Grip Tester for precise pavement friction measurements, and On-Board Diagnostics (OBD) systems to assess vehicle dynamics on the pavement surface. Key roadway safety factors under adverse conditions identified include ice coverage, water film depth, and surface friction levels. To further understand the relationship between these factors and safety, machine learning algorithms—specifically, random forest and gradient boosting algorithms—were applied to identify the most significant variables influencing roadway safety, and corresponding models were developed to assess risk levels for pavement sections under inclement weather. This study provides valuable insights into how pavement and weather conditions impact roadway safety and offers guidance for risk management strategies and roadway design improvements suited for adverse weather conditions.

Keywords: Roadway safety, Pavement conditions, Inclement weather impacts, Machine learning models