Congestion reduction via personalized incentives

The purpose of this research is to develop real-time algorithms to reduce
traffic congestion and improve routing efficiency via offering
personalized incentives to drivers. The incentives and alternative routes
should be chosen smartly in order to maximize the probability of
acceptance by drivers and to avoid the creation of new congestion in other
areas of the network. To this end, we propose to exploit the
wide-accessibility of smart communication devices and develop a real-time
look-ahead incentive offering mechanism using individuals’ routing and
aggregate traffic information. The proposed approach relies on historical
data and state-of-the-art traffic prediction methodologies to continually
predict congestion and traffic flow of the network. Using this prediction
and based on individual preferences, the central controller offers
personalized incentives to drivers with the goal of reducing the
probability of congestion. The decisions about incentives are made via
solving a series of carefully designed large-scale stochastic optimization
problems. The performance of the proposed algorithms are evaluated using
data from the Los Angeles area. Finally, we evaluate the performance of
our method using data from the Los Angeles area. The Los Angeles region is
ideally suited for being the validation area since there are a number of
dedicated carpool lanes in the region and furthermore, there are portions
of the freeway network where congestion pricing is employed with the added
feature that ridesharing vehicles can travel on these lanes free of charge
(e.g., I-110). Additionally, researchers at USC have developed the
Archived Data Management System (ADMS) that collects, archives, and
integrates a variety of transportation datasets from Los Angeles, Orange,
San Bernardino, Riverside, and Ventura Counties.