Snow removal activities are performed by roadway agencies to enhance winter mobility and safety. Slower travel speeds during these operations, combined with low visibility and reduced pavement friction, mean that safety and collision avoidance remain a persistent concern. Many studies have implemented signing and lighting technologies to improve the visibility of snowplows. Although a few studies have evaluated the use of different colors on snowplows, there is no rigorous study that evaluates the potential impacts of using green warning lights for winter maintenance operations. This study, therefore, investigates the impacts of various warning light configurations on the visibility of snowplows, with the focus on green lights. To this end, 37 warning light configurations are designed using various color combinations (green and amber), and flashing patterns (single and quad) on the back (LED), the top (beacon), or both, of snowplows. These configurations are evaluated to identify the most effective configurations. Three sets of experiments are designed and implemented: static, dynamic, and weather to evaluate the visibility effectiveness in different contexts: day versus night, clear versus snowy weather, and static versus dynamic scenarios. Human subjects are employed to conduct the experiments and the test results are evaluated using statistical analyses. The conspicuity during the day time and glare during the night time are statistically different among various configurations. In addition, adding green lights with a single flash pattern to amber warning lights improves the conspicuity, while keeping the glare at an acceptable level relative to configurations using only amber.
Pavement Friction Prediction Using 3D Texture Parameters
