Smart Urban Signal Networks: Initial Application of the SURTRAC Adaptive Traffic Signal Control System

Smith, Stephen F.; Barlow, Gregory J.; Xie, Xiaofeng; Rubinstein, Zachary B. · 2013 · OpenAlex-citations

DOI: 10.1609/icaps.v23i1.13594

archive: archived pipeline: cataloged verified

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

Summary

This paper presents the initial field application of SURTRAC (Scalable Urban Traffic Control), a decentralized, adaptive traffic signal control system designed to reduce urban congestion. The research addresses the limitations of conventional signal timing, which relies on offline plans that cannot respond to real-time traffic variations, and centralized adaptive systems, which often face scalability and computational trade-offs. SURTRAC formulates traffic control as a decentralized, schedule-driven process where each intersection independently computes optimal schedules for approaching vehicles using a single-machine scheduling approach. This method allows for second-by-second responsiveness and network coordination through the exchange of projected outflow data between neighboring intersections. The study details a pilot implementation of SURTRAC on a nine-intersection road network in the East Liberty neighborhood of Pittsburgh, Pennsylvania. The system architecture is a multi-agent model where each intersection runs an embedded agent comprising services for communication, detection, execution, and scheduling. The pilot utilized Trafigon video detection systems to sense vehicle presence and counts, with detectors placed at stop-bars and advance zones to maximize the look-ahead horizon. To handle real-world uncertainties such as sensor errors and communication failures, the system incorporated mechanisms like moving average forecasts for missing data and elasticity measures for queue clearance management. The experimental design involved comparing SURTRAC’s performance against the existing signal control scheme, which used coordinated offline timing plans for rush hours and simple actuated control for other periods. Performance evaluation was conducted through "before" and "after" drive-through runs across four daily periods: AM rush, mid-day, PM rush, and evening. Metrics included travel time, speed, number of stops, wait time, fuel consumption, and emissions. The results demonstrated that SURTRAC achieved significant performance improvements across all studied metrics, ranging from 20% to 40% overall. Specifically, the system reduced travel times and vehicle emissions compared to the pre-existing signal timings. The decentralized nature of SURTRAC allowed it to scale effectively without requiring centralized coordination, while its asynchronous communication protocol proved robust against temporary network failures. The significance of this work lies in demonstrating the viability of decentralized, schedule-driven adaptive control for urban grid networks, which often feature competing traffic flows and densely spaced intersections. By reformulating the optimization problem as a tractable scheduling task, SURTRAC achieves near-optimal timing plans with high computational efficiency. The pilot results validate the system's ability to provide real-time responsiveness and coordinated network behavior, offering a promising alternative to both fixed-timing and centralized adaptive systems for reducing congestion and environmental impact in urban areas.

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.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

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

• vru facing ehmi