Congestion Reduction via Personalized Incentives

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).