IE 11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

New Tools, AI Help Transportation Officials Better Understand Traffic Patterns

Departments of transportation in Massachusetts and Pima County, Ariz., are exploring technology that connects them more closely to their traffic signals.

Two public agencies in opposite regions of the United States are using the same intelligent software and devices to resolve unique transportation needs around maintenance, construction and, above all, communication.

In urban New England, the Massachusetts Department of Transportation (MassDOT) realized the agency needed a more efficient way to keep watch over thousands of state-owned signals that might “come into flash,” or cease to function properly, and to monitor its busier highways, according to traffic engineer Jeff Gomes.

In Pima County, Ariz., the state’s second-most populous county, Michelle Montagnino, a civil engineer at the Department of Transportation, said officials were looking for a better way to move vehicles through construction zones, and to circle in one remote signal where distance limited communication.

Both jurisdictions are piloting solutions from Canadian analytics company Miovision, using its TrafficLink platform, portal and SmartView cameras to improve their on-scene vision and communication, as well as its SmartSense artificial intelligence to do more and better analysis.

The firm, which works with governments and agencies at all levels, aims to unify multiple public-sector “widget” solutions, CEO Kurtis McBride said, similar to the way smartphones aggregated everything from calculators to pedometers.

At MassDOT, the agency began its TrafficLink pilot in Revere, Mass., as part of a Federal Highway Administration pilot, to monitor travel times and do a better job of watching signal performance. The agency installed TrafficLink in four locations, three along a single corridor to enable a deep dive on travel time. Later, it relocated several north to Chelmsford, farther north, to monitor ramps along Route 110.

Overall, the effort allowed MassDOT to control signal timing remotely and scrutinize issues from a central location. It cut so-called “reduced split failures” — when a group of vehicles needs more than one green light to clear an intersection — by 44 percent, the company said.

It “allows us to be able to have a more thorough grasp into what’s actually happening into the intersection,” said Gomes, who added that he sees the need for these types of state-level solutions growing.

“If we had these types of boxes in a lot of intersections where we just have issues that come up based on traffic, it would be a lot easier to diagnose these issues and just have a history we can look back on,” he said. The agency has since wrapped its trial, keeping the devices installed while the company applies to be on the state's Qualified Product List.

McBride said it’s his experience that agencies already have the information they require to act, but they typically need better access and a more comprehensive understanding of what’s happening on their roadways.

“What I always say to cities is, all the data they need to solve the traffic problems they have — they already have the data, it’s just trapped in a metal box on the side of the road. That box hasn’t really changed much in 30 to 40 years, and we’re trying to bring a fresh approach,” McBride said.

He described TrafficLink as a device that “basically provides communication” and reports from the field. SmartSense, he said, is analogous to a “roadside super computer." It completes TrafficLink, offering artificial intelligence capable of mining data from onsite video cameras to present a clearer picture of vehicle behavior and what happens when signals lose power.

“There’s a direct correlation between the amount of stopping that happens in a corridor and the amount of rear-end collisions, [and] between vehicle damage and fatality collisions. We work a lot with our customers to help them quantify that,” McBride said.

In Pima County, the agency controls about 105 signals and about 19 “hawks,” or pedestrian-activated crossing beacons. The hawks aren’t tied into its wireless system, and before standing up its Miovision pilot last year, the agency was able to examine signal controllers and make remote changes — but communication wasn’t always as robust as it could have been.

One of those roughly 105 signals wasn’t connected to the network at all, and so Pima County began using Miovision there, to monitor the intersection and obtain performance data.

It ultimately obtained five Miovision units, installing them in an area which is undergoing a multi-year reconstruction.

Traffic has also been affected by another project to the north, Montagnino said — so the tools, which includes county log-ins to view data graphically on the back end, have helped the agency understand exactly what’s happening on its highways.

“The travel time on Ina [Road] itself did not change, but we’ve had improvements on the north-south [streets]. But it hasn’t been at the risk or at the detriment of the Ina Road corridor,” Montagnino said recently.

By accomplishing traffic observation remotely, the pilot freed up staffers who otherwise would have coordinated those efforts on-scene, and enabled a more extensive look at problem areas. Pima County is currently looking at adding more of the devices, the civil engineer said.

“It would be nice to have more of a network. And so, that’s kind of where we’re going with it, moving forward, is to try to create more of a mesh network with it so we can get some better origin-destination information and then get, obviously, travel times on other corridors as well,” Montagnino said.

Theo Douglas is assistant managing editor for Industry Insider — California, and before that was a staff writer for Government Technology. His reporting experience includes covering municipal, county and state governments, business and breaking news. He has a Bachelor's degree in Newspaper Journalism and a Master's in History, both from California State University, Long Beach.