Skip to main content
eScholarship
Open Access Publications from the University of California

UC Berkeley

UC Berkeley Electronic Theses and Dissertations bannerUC Berkeley

Synchronizing Traffic Signals on Grid Networks During a Rush

No data is associated with this publication.
Abstract

A simple strategy is offered for coordinating all traffic signals on street networks arranged as grids. The strategy is based on the finding that signals of symmetric networks can be perfectly synchronized in two orthogonal directions. With this in mind, the strategy synchronizes signals in the directions that point towards the center of gravity of all workplaces in the morning rush, and away from that center during the evening. Although the strategy was initially developed for symmetric networks, it can readily be adapted to accommodate asymmetrical grid arrangements as well. The strategy is also dynamic and adaptive to real-time traffic conditions. Signals are synchronized according to the free flow motion of vehicles in under-saturated traffic, and to backward waves once street links are filled with queues. An algorithm is developed to switch smoothly between the two synchronization modes as conditions change during a rush.

The strategy was tested via simulations of 20 x 20 signalized grids, operating in non-steady-state traffic conditions typical of rush periods, and benchmarked against a fixed zero-offset strategy commonly recommended for congested traffic. In a scenario representing a morning rush with a monocentric distribution of workplaces, the strategy was also benchmarked against the well-known SYNCHRO program. While SYNCHRO reduced the morning commute delays by 7\% relative to the fixed zero-offset strategy, the proposed strategy reduced delays by 32\% (more than four times that achieved by SYNCHRO). The strategy continued to perform well when workplaces were distributed more broadly over the network, and when workplaces were distributed in a multi-centric pattern. The strategy also performed well when tested for an evening rush, and was adapted to perform well when the grid’s original symmetry was distorted to represent a more realistic transportation network.

Main Content

This item is under embargo until October 30, 2024.