A state DoT in North East manages one of the most dangerous highways in the USA.

The motorway is over 100 miles long, and serves roughly 60,000 journeys per day.

The highway saw over 2,000 crashes in a 3-year period and wanted to increase safety by improving risk detection and expediting incident response. The DoT had 100 cameras throughout the highway, but was looking for a solution that would give it complete coverage of the road, to allow detection of risks along the entire corridor.

“We now have a clearer picture of what's happening on our roadways in a much timelier fashion.”
– TMC manager, а US DоT

Using Lanternn by Valerann™, the TII team were able to detect up to 40% additional true and relevant safety events per day. 50% of events detected, were detected by Lanternn on average 25 minutes earlier than by using traditional methods. Importantly, Valerann provided equitable improvements in operations in areas with and in areas without CCTV coverage.

The DoT together with Valerann deployed Lanternn by Valerann™ into the district’s Control Centre. The system monitored the entire 100-mile corridor. The system integrated with the CCTV cameras, Waze, HERE and Google maps, as well as weather data.

The system used machine vision, data fusion and AI to sift through over 20,000 data points a day and surface only the 50 most relevant events to the control center. The alerts were coupled with contextual data and historical video footage for rapid validation. The system was compared to other incident detection capabilities, to estimate both accuracy and ability to detect novel incidents.