Why Passive Work Zone Warnings Keep Failing

Omnisight CEO Chris McGarty gave the opening partner demonstration at the ITS New Mexico Annual Meeting in Albuquerque on October 14. His topic was work zone intrusions. His first slide was a question: why are the passive systems failing?

Four reasons drivers still hit the work zone

The talk opened with the failure modes, not the fix. Four of them explain most intrusions:

1. Lack of attention. With so much competing for a driver's eyes, people stop registering signs and warnings, or never see them at all.
2. Mobile and temporary operations. Moving work zones are hard to warn for. Sometimes the slow-moving vehicle is the only warning there is.
3. Improper deployment. TMA spacing that is too tight, merge warnings that come too late, no flagger where one was needed.
4. Road geometry. Gore points, blind curves and short merges make some locations more dangerous than any sign can fix.

None of those are solved by adding another sign. Cones, arrow boards and attenuators were designed to survive a crash. They were never designed to see one coming.

The two-second problem

Once a driver fails to merge, the crew behind the taper has roughly two seconds. That is the window every work zone safety tool has to work inside.

• 42,000+ people injured in U.S. work zone crashes in 2024
• 2 seconds for a crew to react once a driver fails to merge
• Under 100 milliseconds from detection to alert with FusionBLADE, processed on the device

What a warning before impact looks like

FusionBLADE fuses HD3D radar with AI video on the back of the attenuator truck. It picks up every approaching vehicle from well beyond the taper, judges whether the driver is responding or on a collision path, and fires the lights and horn at that driver while the crew is warned in the same instant. The alert goes out before the vehicle ever enters the work zone. Every event is recorded, so each closure also becomes a traffic record.

How far out it fires is the agency's call. Omnisight's recommendation, settled over a year of field work with the Missouri Department of Transportation, is 1.5 to 2 seconds before predicted impact. Enough time for the driver to react and avoid the hit. Not so early that crews learn to tune it out.

Proof from Missouri

Chris closed with field clips from MoDOT trucks: a tractor-trailer closing on an attenuator in daylight, a vehicle passing an active scene in the rain at 2:39 in the morning. In each one the system tracked the approach, fired at the tuned threshold and saved the clip. The strike didn't happen. That joint development led to a sole-source award and a statewide attenuator fleet rollout. Read how it started.

Three ways to put it on your trucks

1. Start with the attenuator truck. The TMA is the vehicle that takes the strike, so it is where FusionBLADE goes first. One device on the back watches the approach, fires the lights and horn, warns the crew, and logs the event. That is the setup MoDOT is rolling out across its fleet.
2. Cover the trucks that aren't TMAs with the hitch mount. Striping, patching, mowing and supervisor pickups are the mobile and short-duration operations that are hardest to warn for. The FusionBLADE hitch mount drops into a standard receiver in minutes, runs off the truck or a battery, and moves to whichever vehicle is heading out that day. No permanent modification, no dedicated truck required.
3. Keep the near-miss data. Every approach, alert and close call is recorded on the device. Over a season that record shows which closures, times of day and setups draw the most intrusions, and it feeds the next traffic control plan. The clips Chris showed in Albuquerque came from exactly that log.

Want it on your trucks? Request a demo.

The impact shouldn't be the warning.

Thank you to ITS New Mexico for hosting, and to everyone who stayed after the session to talk through their own work zones with us. See you next year.