Evaluation of weather-related freeway car-following behavior using the SHRP2 naturalistic driving study database

Adverse weather conditions negatively impact the safety, mobility, and reliability of the transportation network. Weather-responsive traffic management (WRTM) strategies have been developed to counteract the hazards of adverse weather on the transportation system. In order to secure investments into applications designed to mitigate adverse weather impacts, explicit evidence of the perceived benefits and challenges are required. Microsimulation modeling is a common tool used to anticipate these impacts and can be used in real-time operational strategies. To achieve reliable output results, realistic driving behavior must be represented in the models. The purpose of this paper is to present findings related to intra-driver heterogeneity as a function of weather: that is, the adjustment of drivers' behavior to compensate for different adverse weather conditions. This study evaluates drivers' car-following behavior in various adverse weather conditions-rain, snow, and fog-and calibrates the Gipps car-following model to identify the transferability of those behavioral nuances. The results produce conclusive evidence that intra-driver heterogeneity exists in different adverse weather conditions and indicate that this heterogeneity can be captured by calibrated Gipps parameters. This study supports the need for the inclusion of weather considerations in current microsimulation modeling practices and contributes a novel trajectory validation methodology that can be used to compare observed and modeled behaviors when correlation between model parameters exists.