Investigating Impact of Sources of Non-recurrent Congestion on Freeway Facilities

Traffic congestion is a growing problem for the transportation system. From the user perspective, there is an increasing need for reliable travel time. Travel time unreliability is the direct result of the variable and often unpredictable events that occur in the transportation system. Different sources can contribute to travel time variability. Weather and environmental conditions, traffic incidents and work zones are important sources of nonrecurrent congestion. The objective of this study is to evaluate the impact of non-recurrent sources of congestion on freeway operations using a fused database approach. Multiple models are evaluated under different operating conditions and recommendations are made based on their goodness of fit to empirical data. The research indicates that inclement weather conditions reduce both freeway capacity and free-flow speed. No evidence of reduction of free flow speed due to incidents was noted. From a traffic stream modeling perspective, the modified Greenshields model provided the best estimate of the freeway traffic stream behavior in almost all the tested conditions. The study recommends estimating free flow speed by fitting the modified Greenshields model and using an empirical threshold approach to estimate freeway capacity. The research findings have been incorporated into existing freeway practice in the Highway Capacity Manual. Finally by applying the recommended traffic stream parameters to an actual freeway facility, the resulting model fit is substantially improved.