College research aims to stop GA traffic jams before they start during emergencies

Anyone who lives in or near Atlanta has been intimately acquainted with traffic jams more times than they’d like to recall. On January 28, 2014, what is usually an infuriating annoyance became a life-or-death situation. An afternoon snowstorm blew through, draping two inches of snow and ice across the city and bringing it to a complete standstill. People were trapped in their cars overnight, in the freezing cold, on every major roadway.

Emergencies like “Snowmageddon 2014” show how susceptible urban roads are to gridlock, or what Georgia Gwinnett College (GGC) Assistant Professor of physics Skanda Vivek calls “cascading road interruptions.” It’s a rather poetic way to describe something that can be very dangerous. 

Vivek is working with undergraduate student Hannah Connor, a senior majoring in mathematics, to show how real-time traffic data can be a powerful, actionable tool to make critical decisions when time is of the essence.

A program serving as Connor’s senior capstone project explores the possibilities of using real-time traffic data not available in the past to inform citizens and governments how to respond to exceptional situations, like snowstorms and hurricanes, or even large-scale civil unrest.

“This research has taught me a traffic network is like a living thing. When something breaks, it finds ways to compensate,” said Connor. “I think this research is important because it could be used to help in emergency situations – for example, helping first responders find the quickest route and avoid potential new or undetected traffic jams.”

The variety of things that can affect traffic flow would surprise most people, said Vivek. 

Using microscopic traffic simulations, Vivek developed a theoretical framework for predicting the growth in cascading traffic jams around disruptions. Vivek found that targeted disruption of roads leads to a disproportionately large proportion of shortest time routes being blocked, due to the relative importance of a few key roads. In other words, it doesn’t take much of a disruption to create gridlock. How many times have you been stuck in bumper-to-bumper traffic for miles, only to finally be set free at the site of a minor fender-bender?

Through a case study in the city of Boston along with previously obtained GPS vehicle trips data, Vivek showed how a targeted attack on the top 1% of intersections would block 40% of shortest-time routes.

He used that data to present at the international conference of the American Physical Society’s March 2021 meeting, where he noted that, aside from natural disasters, the rise of internet-connected vehicles and smart city infrastructure leads to the potential for hackers and nation-states to create targeted disruptions that could maximize the potential for cascading failure.

Developing methods to quantify and mitigate the impacts of cyberattacks on vehicles and intelligent infrastructures in road traffic networks will be essential, said Vivek. Currently, Connor is developing algorithms through multidimensional data sources, including the HERE traffic API, to detect signatures of large-scale anomalies on networks, which can divert drivers in real-time to keep traffic flowing. Vivek and Connor’s work was highlighted by the official HERE technologies blog

The work of the GGC team, ideally, will lead to safer, more efficient roads. It’s serious business, said Vivek, because ultimately, it won’t just save time – it will save lives.

Information from Georgia Gwinnett College.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Previous Story

Carr: Coalition of Attorneys General Urges Biden to Abandon Court-Packing Scheme

Next Story

Georgia Man indicted for laundering over $2 Million dollars in Email scheme

Sign Up For Our  Newsletter
Get the latest headlines and stories - and even exclusive content!- sent right to your inbox.
Stay Updated
Give it a try, you can unsubscribe anytime.