Rapid Flow Blog

Portland Reduces Delays by 20% at Maine’s Busiest Traffic Intersection After Surtrac Deployment

Posted on February 14, 2019

After deploying Surtrac, Rapid Flow Technologies’ real-time, artificial intelligence (AI) adaptive traffic signal control system, the city of Portland experienced a 20% reduction in delays and a 16% reduction in travel time through the “Morrill’s Corner” intersection system.

Morrill’s Corner includes the busiest intersection in Maine where 33,000 vehicles pass through each day, according to a 2016 report from the Maine Department of Transportation. That means that on average, travelers through this intersection are saving 9,350 minutes, or almost 156 hours, each business day and nearly 41,000 hours per year[1]. This time savings results in lower direct costs (e.g. time and fuel) as well as indirect costs (e.g. emissions, safety, and business/freight efficiency).

But one of the most important KPI’s (key performance indicator) for an initiative like this is customer experience, and so far, the feedback has been overwhelmingly positive. Dan Pratt, a Metro bus driver who drives through Morrill’s Corner on his route, was quoted in a recent Portland Press Herald article saying, “It has shorted up the wait times a lot.”  Julie Rosenbach, a local commuter said, “I’ve noticed a big difference…it used to back up a few blocks.”

Portland’s Surtrac deployment, as well as other “smart city” investments to maintain and improve quality of life, were partly funded by a program to replace 6,000 of Portland’s old street lights with a state-of-the-art LED network. In addition to implementing free public Wifi, Portland plans to use the savings from the LED project to expand its Surtrac deployment in other neighborhoods around the city. Expanded intersections could include those in the downtown area where tourists, many from visiting cruise ships, create not just vehicle congestion issues, but pedestrian congestion and safety issues as well.

How Surtrac Works and Why It’s Effective

While Surtrac works great in more predictable suburban arterial environments, it is ideal for more complex networks, including grids and closely spaced intersections, where competing traffic flows require service from different directions or where flows are unpredictable and change in ways that makes time-of-day patterns obsolete very quickly. These are the very conditions present at Morrill’s Corner, and why Surtrac intelligent traffic signal control has been so effective in reducing congestion there. Further, there is no doubt that as Portland's population continues to grow, the traffic conditions at Morrill’s Corner will evolve and change over time. Surtrac naturally adapts to these changes using its second-by-second AI optimization.

Traffic at Morrill's Corner in Portland with map

Surtrac’s patented, artificial intelligence optimization engine, is a decentralized, highly flexible, edge-computing system that allows each intersection to operate independently, or to coordinate with its neighbors, depending on actual traffic demands. This theory of independence AND coordination makes it different from almost every other adaptive traffic control system available. Further, the Surtrac adaptive traffic management system optimizes in real-time – literally every second – allowing it to adapt quickly to changes in traffic patterns. Because Surtrac is decentralized, there is no latency and it can scale to networks of any size without adding network compute complexity. Decentralization also comes in handy for a city like Portland who plans to add groups of intersections over time, versus all at once.

How Surtrac Helps Cities Adapt to Changing Current & Future Transportation Demands

Though pedestrian traffic at Morrill’s Corner is quite low and is not a significant issue, this is NOT the case for most of Surtrac’s other deployments, and certainly not the case for most of the other neighborhoods in Portland – especially the heavy tourist sections in downtown near the harbor. 

Surtrac is a smart city traffic system designed to be truly multi-modal. So long as the detection system can identify different types of travelers, Surtrac can consider them in its optimization planning. Any mode of transportation can be accommodated and/or prioritized, including but not limited to pedestrians, cars, delivery trucks, transit vehicles, scooters, motorcycles, etc. In fact, Surtrac can be used just as effectively at pedestrian-only intersections and crossings (often referred to as “Hawks”). And because the system is decentralized, different intersections can have different configurations based on their unique needs. 

Surtrac is also designed for an environment that includes connected and autonomous vehicles (CAV). In addition to communicating among intersections, Surtrac takes an I2X (intersection to everything) approach to communications. It can receive information from, and can send information to, vehicles, pedestrians, and other ITS systems and sensors, among others. One of the most exciting recent communication enhancements is the ability for disabled or older pedestrians to request more crossing time directly to the Surtrac system via a handheld app (though this app is still only in prototype or “alpha” phase).

It is this type of multi-modal and CAV/communications flexibility that is important to Portland as it expands Surtrac into neighborhoods with high pedestrian traffic, and as it relies on Surtrac to support future transportation realities that are just emerging today.

 


[1] Surtrac was measured to save each traveler an average of 17 seconds when traveling through the three intersection section of Morrill’s Corner. When multiplied against 33,000 average travelers per day, that results in an average savings of 561,000 total seconds, or 9,350 minutes, or 155.83 hours per day. Multiplied by 261 business days, this results in 40,716 hours saved per year.

This entry was posted in Case Studies.
surtrac-logo
Grab the Product Sheet
The Quickest Way to Learn About Surtrac

To Talk More About Improving Traffic in Your City

Contact Us

Recent Posts