The Price of Anarchy in Transportation Networks: Data-Driven Evaluation and Reduction Strategies

Abstract

Among the many functions a Smart City must support, transportation dominates in terms of resource consumption, strain on the environment, and frustration of its citizens. We study transportation networks under two different routing policies, the commonly assumed selfish user-centric routing policy and a socially-optimal system-centric one. We consider a performance metric of efficiency -the Price of Anarchy (PoA) -defined as the ratio of the total travel latency cost under selfish routing over the corresponding quantity under socially-optimal routing. We develop a data-driven approach to estimate the PoA, which we subsequently use to conduct a case study using extensive actual traffic data from the Eastern Massachusetts road network. To estimate the PoA, our approach learns from data a complete model of the transportation network, including origin-destination demand and user preferences. We leverage this model to propose possible strategies to reduce the PoA and increase efficiency.

Learning and Management for Internet of Things: Accounting for Adaptivity and Scalability

“…Age of information is pertinent to mission-critical IoT applications [28], [88]. Furthermore, for traffic assignment tasks in intelligent transportation, f t (x t ) can capture the overall fuel consumption, and the travel time of vehicles on the road [106]; for demand response in smart grids, it is related to user utility and power balancing cost depending on the real-time energy prices [52], [54], [85], [97]; and for applications related to wireless communications, throughput or achievable rate also plays a critical role in the objective. Short-term constraints.…”

“…, selecting x t ∈ X (s t ) guarantees the E2E latency requirement in MEC. Short-term constraints also arise due to the physical limits of transmission lines and generators in power networks [40], transceivers in wireless communication [96], as well as vehicles in transportation networks [106]. Long-term constraints.…”

The Penetration Rate Effect of Connected and Automated Vehicles in Mixed Traffic Routing

“…In particular, we seek to find the route occupancy matrix (probabilities) for allocating vehicles to routes. In other words, find the probability matrix P where its elements denote the probability that a vehicle traveling from an origin O to destination D uses route r. 1) Problem Formulation: As in [13], we model the traffic network as a directed graph G = (V , A , W ) where V is the set of nodes, A is the set of links, and W = {w i :…”

“…6b) is chosen for the case study. For this network, we use the O-D demand which has been estimated using an inverse optimization framework in [13]. In order to solve the problem we consider 56 O-D pairs, and allow up to 3 routes between each origin and destination (top 3 shortest routes).…”

“…Braess network routing results under different penetration rates by solving system-centric time-optimal problem (a), (c), (e) and, system-centric energy-optimal problem (b), (d), (f). (a) All available road segments in the road map of Eastern Massachusetts[13] ; (b) Interstate highway sub-network of eastern Massachusetts and a simple illustrative example were used to validate the models.…”