Busy Arizona Highway Uses Intelligent IoT to Ease Gridlock

Smart transportation technology helps tap the breaks on traffic along Arizona’s Bell Road Highway. GRIDSMART uses real-time data and edge computing to improve travel time and intersection efficiency, even when the tourists come.


Article Key

image credit: GRIDSMART SMARTMOUNT Bell Camera

No one likes traffic. It’s unavoidable in populated areas, but having the ability to reduce congestion can improve the quality of life for citizens and save wear and tear on roads. Civic leaders in Arizona were battling traffic problems in one particular stretch of road and needed a solution that enabled them to improve vehicle flow.

Bell Road Highway in Surprise, Arizona, sits about 25 miles north of Phoenix. On average, 75,000 vehicles traverse the six-lane highway every day, about 30 percent more capacity than a typical six-lane highway. As a result, traffic overwhelmed about half of the 31-mile road, especially on busy days, such as when Major League Baseball spring training events brought in nearly 2 million fans.

The city, along with six other agencies, looked into widening roads to accommodate the extra traffic, but cost estimations ran as high as $24 million, plus an annual operating cost of $100,000. Instead the group sought an intelligent technology solution, which was estimated at $1.6 million, plus annual operating costs ranging between $30,000 to $50,000.

The pilot project tested adaptive signal control technology in multiple areas and analyzed traffic efficiency during spring training. The solution had to improve traffic flow and safety, adapt for changing traffic patterns, and provide signal timing progression. It also had to accommodate event-based traffic increases, seasonal traffic increases, and unexpected occurrences, such as accidents or lane closures. The project would encompass 50 traffic lights on Bell Road and would focus on nearby freeway interchanges.

Traffic Management on the Edge

The group chose the GRIDSMART intelligent traffic monitoring system, which uses Intel® Distribution of OpenVINO™ toolkit as part of its vision processing engine and AI. The system uses computer vision tracking algorithms to follow moving objects in the intersection. It collects and analyzes real-time traffic data, enabling civil engineers to adjust traffic signals based on usage, thus improving vehicular flow and safety.

GRIDSMART is comprised of three elements: the SMARTMOUNT Bell Camera, the GS2 Processor, and the GRIDSMART Client software. The five-pound Bell camera contains a protected fisheye lens that provides a full view of the intersection. The expansive view allows the camera to capture turn counts, provides situational awareness, views of the center of the intersection and the roads feeding into it, and incident management views. It operates via a Power over Ethernet connection from the GS2 processor.

The data from the camera is sent to the GS2 Processor, where the edge computing takes place. Running the GRIDSMART Engine, a suite of proprietary vision-tracking algorithms, the GS2 processes the camera images to build a 3D model of all vehicles, people, and other objects that enter each intersection. The resulting data is sent to the cloud every night and shared with all controllers. The GS2 is contained in a 1U-high enclosure that can be rack-mounted or stand alone. The front panel LEDs indicate phases, calls, and signal status.

The system is managed by the GRIDSMART Client. Traffic engineers can use the virtual pan-tilt-zoom functionality to configure, view, and manage intersections and highways in real time, and the virtual DVR capability allows engineers to monitor each intersection at different angles and speeds. Engineers can set up detection and counting zones, and receive automatic alerts, notifications, and reports via email. The GRIDSMART system can also be used with other traffic management systems.

Power in Numbers

In the Bell Road Highway implementation, GRIDSMART handled detection and data collection, and the data was fed into the Kadence advanced traffic management system from Kimley Horn. Using its algorithms, Kadence suggested new cycle lengths, splits, timing plans, and other adaptive controls to improve traffic flow.

Multiple views of traffic

Image credit: GRIDSMART Client

The GRIDSMART and Kadence solution boosted travel times and intersection efficiency as evidenced by the numbers. In the test area, delays were reduced by 20 percent on weekdays and 43 percent on weekends. Travel times were cut by 2 percent every day, and overall speed increased nearly 2 percent. Even with the spring training tourists, delays were reduced by nearly 39 percent, travel times dropped 5.5 percent, and overall traffic speed increased 5 percent. According to calculations by the Arizona Department of Transportation, that translates to a savings of $9,500 each day and $1,437.36 on days of special events.

Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries.

Find Intel® IoT Market Ready Solutions

Search the Intel® IoT partner ecosystem and discover 130+ end-to-end IoT solutions.

Become a Member of the Intel® Partner Alliance

If you are delivering repeatable and scalable IoT offerings using Intel technology, the Intel® Partner Alliance program gives you more opportunities to build your business. Get started.