Empirical study of impact of icy roadway surface condition on driver car-following behavior

Research was done to quantify the impact of icy roadway conditions on driver car-following behavior. The data used in the study were gathered by a group of researchers at Japan's Hokkaido University in a controlled environment under dry and icy roadway conditions. The data were used to calibrate the Van Aerde steady-state car-following model, along with vehicle acceleration and deceleration constraints. The impact of icy roadway conditions on five driver-specific car-following parameters [driver perception-reaction time (PRT), free-flow or desired speed, speed-at-capacity, capacity, and jam density] was conducted with one-way analysis of variance and Kruskal-Wallis tests. The results demonstrate that icy roadway conditions produce statistically significant differences at a 5% significance level. Specifically, icy roadway conditions reduce the mean free-flow speed, speed at capacity, and capacity by 28%, 13%, and 46%, respectively, compared with driving on dry roadways. The mean PRT for icy conditions is found to take 13% longer than driving under dry conditions. The longer PRTs can be attributed to drivers driving at lower speeds and larger spacing on icy surfaces compared with dry conditions.