Search

With rapid population growth and urban development, traffic congestion has become an inescapable issue. Future mobility services may offer a new and very effective opportunity to incentivize congestion-reducing behavior. Ridesourcing, food delivery, and package delivery companies make decisions that directly or indirectly affect routing for an increasing number of drivers. These organizations have more flexibility and more power to affect traffic, and targeting them with incentives may be more efficient than incentivizing individual drivers.Motivated by this idea, researchers at the University of Southern California developed a distributed algorithm for offering incentives to organizations to make socially optimal routing decisions. The algorithm is designed to lower the traffic flow of congested roads without creating new congestion in other parts of the road network. This policy brief summarizes the findings from that research and provides research implications.View the NCST Project Webpage

With rapid population growth and urban development, traffic congestion has become an inescapable issue in large metropolitan regions. Research studies have proposed different strategies to control traffic, ranging from roadway expansion to transportation demand management programs. Among these strategies, congestion pricing and incentive offering schemes have been widely studied as reinforcements for traffic control in traditional traffic networks where each driver is a “player” in the network. In such a network, the “selfish” behavior of individual drivers prevents the entire network to reach a socially optimal operation point. In future mobility services, on the other hand, a large portion of drivers/vehicles may be controlled by a small number of companies/organizations. In such a system, offering incentives to organizations can potentially be much more effective in reducing traffic congestion rather than offering incentives directly to drivers. This research project studies the problem of offering incentives to organizations to change the behavior of their individual drivers (or individuals using their organization’s services). The incentives are offered to each organization based on their aggregated travel time loss across all their drivers. This step requires solving a large-scale optimization problem to minimize the system-level travel time. We propose an efficient algorithm for solving this optimization problem. To evaluate the performance of the proposed algorithm, multiple experiments are conducted by Los Angeles traffic data. Our experiments show that the proposed algorithm can decrease the system-level travel time by up to 6.9%. Moreover, our experiments demonstrate that incentivizing organizations can be up to 8 times more efficient than incentivizing individual drivers in terms of incentivization monetary cost.View the NCST Project Webpage

Rapid population growth and development in cities across the globe have fueled an inescapable urban burden: traffic congestion. Congestion causes huge economic losses and increased vehicle emissions, which contribute to poor air quality. Over the past several decades traffic engineers have tried various strategies to reduce congestion, ranging from increasing roadway capacity to transportation demand management programs.An alternative travel demand management approach that has garnered less attention is the use of positive incentive programs— rewarding desirable behavior rather than penalizing undesirable behavior—to reduce congestion. Researchers at the University of Southern California developed a real-time, distributed algorithm for offering personalized incentives to individual drivers to make socially optimal routing decisions. The methodology relies on online and historical traffic data as well as individual preferences and routing options from drivers’ origins and destinations to estimate both the traffic condition and the drivers’ responses to the provided incentives. This policy brief summarizes the findings from that research and provides policy implications.View the NCST Project Webpage

With rapid population growth and urban development, traffic congestion has become an inescapable issue, especially in large cities. Many congestion reduction strategies have been proposed in the past, ranging from roadway extension to transportation demand management programs. In particular, congestion pricing schemes have been used as negative reinforcements for traffic control. This project studies a different approach of offering positive incentives to drivers to take alternative routes. More specifically, an algorithm is proposed to reduce traffic congestion and improve routing efficiency by offering personalized incentives to drivers. The idea is to use the wide-accessibility of smart communication devices to communicate with drivers and develop a look-ahead incentive offering mechanism using individuals’ routing preferences and aggregate traffic information. The incentives are offered after solving large-scale optimization problems in order to minimize the expected congestion (or minimize the expected carbon emission). Since these massive size optimization problems need to be solved continually in the network, a distributed computational approach is developed where a major computational burden is carried out on the individual drivers' smartphones (and in parallel among drivers). The convergence of the proposed is an established distributed algorithm under a mild set of assumptions (that are verified using real data).View the NCST Project Webpage