Smartphone Based Incentive Framework for Dynamic Network Level Traffic Congestion Management

Du, Lili; Peeta, Srinivas; Ning, Yuqiang; Anne, Viswa Sri Rupa · 2022 · ROSA P / Southeastern Transportation Research, Innovation, Development and Education Center (STRIDE)

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This study addresses the challenge of managing dynamic, network-level traffic congestion by leveraging smartphone-based real-time incentives to influence driver routing decisions. Motivated by the high economic and environmental costs of urban congestion, the research explores demand-side management techniques as a more sustainable alternative to supply-side infrastructure investments. The framework aims to utilize tangible incentives (monetary or value-based credits) and intangible nudges to subtly guide drivers toward less congested routes, thereby improving overall system performance. The research is structured into two primary tasks. Task 1 investigates the generation of real-time incentives using a reinforcement learning (RL) approach. The problem is formulated as a Markov decision process where the state is defined by link travel times and the action consists of incentive values assigned to network links. The reward function balances the reduction of total system travel time against the cost of incentives provided. The Deep Deterministic Policy Gradient (DDPG) algorithm is employed to handle continuous state and action spaces, training the model to optimize incentive distribution. This task also incorporates a conceptual smartphone app framework for disseminating these incentives and modeling user responses, including habitual users and those responsive to nudges. Task 2 focuses on information provision strategies, developing a Correlated Equilibrium Routing Mechanism (CeRM). This mechanism exploits information gaps between individual drivers and a central planner to guide route choices toward a system-optimal equilibrium while respecting individual selfishness. A Distributed Augmented Lagrangian (D-AL) algorithm is developed to solve the CeRM efficiently for real-time navigation services. Numerical experiments and simulations demonstrate the effectiveness of both approaches. In Task 2, simulation results on the Sioux Falls network indicate that the CeRM significantly outperforms existing mechanisms. Specifically, the CeRM reduced system cost by 55% compared to an Independent Routing (IR) mechanism and by 3.6% compared to a user-oriented Equilibrium Routing (uoER) mechanism. The D-AL algorithm proved capable of solving the routing problem quickly, supporting online real-time applications. Task 1’s RL framework successfully generated dynamic incentives responsive to congestion levels, with the conceptual app design providing a pathway for real-world deployment. The significance of this work lies in its contribution to efficient, technology-driven traffic management tools. By integrating behavioral psychology with advanced computational algorithms, the study provides transportation agencies with complementary strategies to mitigate congestion without relying solely on infrastructure expansion. The findings suggest that smartphone-based incentives and correlated routing mechanisms can substantially reduce travel times and system costs. Future work includes testing the framework on real-world networks like Atlanta and incorporating drivers' prior knowledge into the routing models to enhance compliance and effectiveness.

Key finding

The Correlated Equilibrium Routing Mechanism significantly reduces traffic congestion and system travel time by 55% and 3.6% respectively compared to Independent Routing and User-oriented Equilibrium Routing mechanisms.

Methodology

modeling

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• traffic density