This report presents a vehicle reidentification algorithm for consecutive detector stations on a freeway, whereby a vehicle measurement made at a downstream detector station is matched with the vehicle's corresponding measurement at an upstream station. The algorithm should improve freeway surveillance by measuring the actual vehicle travel times; these are simply the differences in the times that each (matched) vehicle arrives to the upstream and downstream stations. Thus, it will be possible to quantify conditions between widely spaced detector stations rather than assuming that the local conditions measured at a single station are representative of an extended link between stations.

Boundary flow control is potentially one of the most effective approaches for city traffic management. Under the control scheme, a city is partitioned into multiple, properly-sized neighborhoods and the transfer flows crossing neighborhood boundaries are metered. In this work, we divide boundary flow control into two stages and address them separately. In the first stage, the total metering rate of each boundary is determined. In the second, the boundary metering rate is distributed among individual road links.

The first stage is handled by a model-predictive controller embedded within a macro- scopic traffic model. The most commonly used macroscopic traffic model, which we refer to as Neighborhood Transmission Model (NTM), uses Macroscopic Fundamental Diagrams (MFDs) in combination with flow conservation laws to model the dynamics of neighborhood vehicle accumulations. We propose an enhanced NTM, which resolves multiple concerns of previous models. Unlike its predecessors, our model accounts for the constraining effects that outbound queues at the boundary can impose on a neighborhood’s circulating traffic. It also differentiates between saturated and under-saturated metering operations. With the proposed NTM, we use a model-predictive control algorithm to generate control actions at small time steps based on their forecasted impacts. Our scalable numerical method can do this for an arbitrary number of neighborhoods within a city. Computer simulations show that these enhancements improve the predictions of trip completion rates in a neighbor- hood, and of the rates at which vehicles cross metered boundaries. Optimal control actions generated as a result are similarly shown to do a better job in reducing the vehicle-hours traveled (VHT) in a city. VHT reductions stemming from the proposed model and from its predecessors differed by as much as 18%.

The second stage of control distributes boundary metering rates onto individual road links along the controlled boundary, which we refer to as the metering distribution. An effective model-free approach for determining the metering distribution is proposed in this work. To do this, we represent the street network as a directed graph and formulate the metering distribution problem as a Reinforcement Learning (RL) problem, which is solved with an actor-critic method. In extensive numerical experiments, the RL-based metering distribution policy shows consistently better performance in terms of VHT reduction than a baseline policy.

This dissertation aims to increase the understanding of supply- and demand-side congestion management strategies. In particular, the work focuses on left-turn bans and congestion pricing. It proves the existence of an optimal spatiotemporal zone within which to ban left turns during the morning rush in cities with central business districts. The optimal ban is capable of reducing vehicle hours travelled significantly. Contrary to popular assumptions, left-turn bans are also capable of reducing vehicle miles traveled. The effectiveness of turn prohibitions is impacted by the distribution of origins and destinations, route lengths and time of day. Moreover, this dissertation explores the effects of combining turn prohibitions with different types of congestion pricing strategies. For the case of Stockholm, simulations show that combining cordon-based congestion pricing with the optimal portion left-turn ban creates a win-win scenario. Either additional travel time benefits are gained at the original toll levels; or the original benefits are maintained at reduced toll levels. Combining the two measures also highlights the importance of optimizing left-turn bans since banning turns in wholesale fashion does not improve Stockholm’s travel conditions. Furthermore, combining turn prohibitions with distance-based congestion pricing for a case study of downtown LA creates synergistic benefits. The reductions in travel costs due to the joint deployment of both strategies is greater than the sum of the reduction due to each strategy on its own. Adding turn prohibitions to a priced network allows commuters to arrive to their destinations both faster and closer to their desired arrival time.

The morning rush in two idealized settings are forecast into the future, assuming that most drivers will engage Adaptive Cruise Control (ACC) while driving their cars. To this end, careful measurements of ACC-equipped cars traveling on roads and a highway confirm an earlier finding reported by others: for a given traffic speed in ACC-rich congestion, the density tends to be smaller than in present-day congestion where all vehicles are manually operated. All else equal, the lower congested densities (i.e., larger car spacings), mean that queues will be less-densely-packed and will expand over longer physical distances in the future, as ACC-equipped vehicles become more prevalent. These uncompacted queues pose problems for urban areas, where queue storage is often already a problem during rush hours.Simulations calibrated using the field-measured data confirm this concern and contradict optimistic predictions of how ACC may lead to a congestion-free future. In a setting with already moderately high congestion and long-street links inspired by Downtown Los Angeles, it is predicted that networkwide vehicle hours traveled (VHT) will increase by up to 12% compared to present-day levels, despite assumptions that are largely favorable to ACC. In a setting inspired by Midtown Manhattan, with short-street links and already high congestion levels, networkwide VHT is predicted to increase by as much as 87%. The higher bottleneck capacities often promised by advocates of ACC are shown to be irrelevant when spillover queues restrict flows from reaching those capacities. Interventions tested through simulations include adjusting onboard ACC controllers to produce lower jam spacings, to prevent a problematic future.

This project presents a signal controller algorithm to capitalize on the extended information provided by wide-area detection at isolated intersections. Using computer simulation, different control strategies are evaluated and the performance of the proposed wide-area detection system with conventional signal controllers is compared. The results indicate that wide-area vehicle actuated (VA) control can yield significant improvements over conventional VA control strategies.

Offered here is a critique of a simple scheme recently proposed for metering freeway on-ramps. An earlier report of this scheme's potential for reducing commuter delay is shown to be exaggerated. The discussion makes clear that to reduce delay, metering should increase the rates at which commuters exit the freeway. The scheme critiqued here, as well as other well-known metering algorithms, are shown to have deficiencies in this, particularly when the freeway is plagued by a diverge bottleneck with a congested off-ramp. Other more effective schemes for reducing the delay caused by these types of bottlenecks are described. Suitable ways of field-testing freeway traffic management strategies are discussed as well.