Optimizing HOT Lane Performance Using Congestion Pricing Algorithms

High-occupancy toll (HOT) lane operations have been implemented in several urban areas in the United States and are considered an effective measure to mitigate demand through congestion pricing. However, literature indicates a lack of predeployment evaluation of congestion pricing algorithms (CPAs) and traffic optimization due to the complexity of dynamic HOT strategies. This paper presents a tool that can predict the potential of a CPA to manage a HOT lane operation in an urban freeway. The authors develop a practically operational feedback-control that not only recognizes drivers’ lane-switching behavioral responses to dynamically changing tolls and traffic conditions but also dynamically optimizes HOT lane operation under a CPA. This would enable forecasting expected HOT lane performance prior to project implementation. Results show that the HOT lane effectively utilizes excess capacities of the HOV lane under the CPA. HOT lane demand nearly doubles while general-purpose lane speeds increase by 25 percent during peak periods, showing an overall improvement in traffic mobility along the freeway. Overall, the developed tool performed reasonably well in optimizing traffic operations of the HOT lane system under various traffic conditions. Further, a sensitivity analysis was conducted by systematically changing model input parameters to determine the effects of such changes on the model predictions.