GridMatrix Launches Next Generation Roadway Safety Analytics Platform
First-of-Its-Kind Safety Analytics & Instant Incident Detection Solution Suite Empowers Cities to Predict & Respond to Multiple Hazards in Real Time. Better Data is the Key.
SAN FRANCISCO, CA, Monday April 10th, 2023 — GridMatrix, the leading provider of cloud-based transportation analytics, is proud to announce that it has launched the next generation of its cloud-based roadway safety platform. This new release includes dozens of new features and improved functionality, empowering users to predict, quantify, and respond to a range of roadway hazards.
These features are available as of today in beta to GridMatrix’s customers, and the company will make them available for public demonstration throughout the spring at:
Intelligent Transportation Society of America’s annual meeting in Dallas, Texas in April 2023
Smart Cities Connect Spring Conference in Denver, Colorado in May 2023
National Association of City Transportation Officials (NACTO) annual meeting in Denver, Colorado in May 2023
These new features include:
- Advanced Filtering for Near Miss Events: Quantifying and sorting near miss events by post post-encroachment time (PET) [1] and time-to-collision (TTC) [2] with the added ability to filter and isolate the worst incidents, especially those including a vulnerable road user (VRU)
- Severity Ratings for Collisions & Near Miss Events: Using near miss data to identify intersection “hotspots”, quantify incident severity, and normalize interactions between all classes of road users
- Identifying “Special Events”: Detection for hazardous “special events” such as rear-ends, disabled vehicles, shoulder-parked vehicles, pedestrians out-of-crosswalk, and more
- Real Time, Custom Alerting: Notification generation for “special events” as well as for custom thresholds for any data fields within GridMatrix’s platform.
More than 43,000 people died on American roads in 2021 and the National Highway Traffic Safety Administration estimates that 2022 will match or even exceed that number.
Traffic engineers, as well as roadway, and facilities operators face the daunting challenge of either monitoring multiple video streams at once to respond to events and have few options when it comes to predicting where the next incident might occur.
The next generation of GridMatrix’s roadway safety platform gives these users the tools to proactively monitor and respond to incidents as well as drive their mitigation decisions with data.
Finding the Signal in the Noise — Using Incident Context Data to Focus on the Near Miss Events that Matter
GridMatrix’s software can now measure (and in turn allow users to filter) near misses by object types, speeds, & severity — all in real time. Our platform can accurately pick out multiple vehicular and pedestrian road user types. Vulnerable road users (VRUs) are broken out by type and include pedestrians and cyclists, while vehicle types include multiple classes of passenger vehicles, fleet vehicles, and trucks. In turn, when a near miss is detected, a time stamped record of the incident is immediately created. Context data including road users types involved, as well as their speed, location, and near miss severity as measured by post encroachment time (PET) or time-to-collision (TTC) is all logged.
Collectively, all this granular information surrounding each event forms a rich picture of an intersection’s safety profile. Filtering allows users to find the signal in the noise — the patterns that matter most. One intersection might have a high rate of near-misses between trucks and people, concentrated at speeds over 50mph, and with less than 2s PETs (a very narrow miss). This is a flashing alarm demanding immediate attention. The high speeds and low PET values already suggest a dangerous situation, but the fact trucks weigh significantly more than pedestrians implies that these near misses could have catastrophic consequences. Now road operators have the tools to understand and mitigate these hazards before they can be realized.
How Bad is it Really and How Do We Know? — Predicting Collisions & Quantifying Near Miss Incident Probability
What do you get when you combine real-time counts of all road users with real-time counts of near misses? (in turn stratified by a rich set of context data about each incident such as object types involved, speed, and location)? Answer: the data and confidence to answer “is this road dangerous really? For the first time ever, GridMatrix’s platform provides municipalities, transportation planners, and traffic engineers have not ever had 24/7 access to these streams of data.
Every day, a range of transit-related special events can impact the healthy operation of an intersection, roadway, bridge, or tunnel. In turn, the disruptions caused by these events have waterfall effects. Sudden breaking, accumulated idling and emissions, and the increased risk of collision all accompany these special events. Fast responses are critical to mitigating these undesirable outcomes and keep roadways functioning smoothly and safely, but how can roadway operators respond if they don’t know what’s happening?
To some extent, incidents can be inferred — significant breaking and speed reductions suggest there may have been an accident. That said, unless someone reports it (or even less likely, a roadway operator happens to catch it on a feed they’re monitoring) the time elapsed magnifies the severity of these waterfall effects. The first step is identifying that such an event has taken place, and confirming what kind of event it is. This basic information differentiates the stopped vehicle in the middle of 5 lanes of freeway traffic from the pedestrian that has walked into a vehicle-only tunnel.
Timing is Everything — Receiving Alerts in Real Time For The Metrics that Matter Most
GridMatrix’s platform now provides user-customizable notifications for all data fields we collect across congestion, signal performance, emissions, and safety. Our users have told us time and again that data is great but insights are better. “Queue lengths at this intersection are 10 cars long — is that a lot?” “How do I know?” Notably, the insights users are looking for are always unique. Facility & roadway operators want to know when there’s a disabled vehicle blocking traffic and creating a bad jam so they can dispatch emergency responders. Toll operators planning for capacity want to know when queues exceed a certain length to know when to open more lanes. Traffic engineers managing intersection signal timing want to know when the proportion of road users arriving on green falls below a specific threshold so they can switch plans.
Since these notifications are user-defined, they can be customized to fit multiple needs. What they all have in common, however, is the ability to leverage the vast pool of historical data GridMatrix collects from daily operation and then to compare average values for a location to the present moment. A queue length of 10 cars might not sound long — unless it’s 5 times the location’s hourly average. Users now have the power to passively monitor their entire road grid and receive active highlights on the flash points that matter most to them.
About GridMatrix
GridMatrix’s award-winning software platform for real-time traffic analytics combines edge data from existing roadway sensors and cloud-based data sources to eliminate urban traffic congestion, accidents, and emissions.
The Intelligent Transportation Society of America has recognized GridMatrix’s software as a leading new solution for sustainable and resilient infrastructure.
GridMatrix is headquartered in California with operations across the United States. The company was founded in 2021 by a group of former Apple engineering and operations colleagues. The company’s recent work has been featured in Bloomberg, the Associated Press, and Government Technology.
For press inquiries, please contact:
Madison Harris
mharris@gridmatrix.com
Term Glossary
[1] Post Encroachment Time (PET) — the time elapsed between when one object’s position intersects with a second object’s previous position
[2] Time to Collision — The time it would take for one vehicle trailing a second to collide with it
[3] Vulnerable Road Users (VRUs) — pedestrians, cyclists, and non-motorized roadway users
