Approaches to Congestion Pricing for High-Occupancy-Toll (HOT) Lanes Using Control Theory
Research Lead: Wenlong Jin
UC Campus(es): UC Irvine
Problem Statement: In the U.S., high-occupancy-vehicle (HOV) lanes are widely used on freeways to reduce congestion since the 1970s. Such lanes are reserved for cars with a minimum of two or three occupants and other qualified vehicles. Under some conditions, however, the HOV lanes are underutilized when the whole freeway system is congested. High-occupancy-toll (HOT) lanes have been one of the most successful lane management methods, which combine HOV lanes and congestion pricing strategies by charging single occupancy vehicles (SOVs) to use HOV lanes during peak periods. HOT lanes offer numerous benefits to both the operators and users by making use of the excess capacity on HOV lanes and have been implemented on State Route 91, US Route 101, and I-15 in California and other places. Generally, the operational goals for the HOT lanes are two-fold: (i) maintaining the freeflow condition; and (ii) maximizing the usage of the HOV lanes. This will help to guarantee the trip time reliability of both HOVs and paid SOVs as well as to minimize the congestion level on the general purpose (GP) lanes. The success of achieving these goals is predicated on the determination of prices dynamically, which in turn depends on an understanding of the values of time (VOTs) of SOVs. Thus, two outstanding problems are to determine a dynamic price for time-dependent demands of both HOVs and SOVs and to estimate SOVs’ VOT.
Project Description: This research will propose two approaches to congestion pricing for HOT lanes using well established techniques from control theory. In the first method, the research team will assume a logit lane group choice model to estimate how drivers will select lanes under specific costs and travel times. The research team will then apply a proportional-integral (PI) controller to estimate the average VOT for SOVs and use this to calculate the dynamic price based on the logit model. In the second method, the research team will implement a dynamic feedback control scheme with a proportional and double integral (𝑃𝑃𝑃𝑃2) controller to determine the dynamic price. Lastly, the research team will show how to estimate the distribution of VOTs indicated by those prices.
Status: In Progress