Connected systems
Connect what you already run, and stop reconstructing it later
An ELD connection and a TMS connection are live. On top of them sit GhostRun, which reports a unit moving with nobody logged into it, and Sentinel, which raises the checks somebody was otherwise meant to remember.
Running now
Earlier today
Read 2 minutes ago. An episode closes itself when a driver logs in.
The ELD already knows. So does the TMS.
Both systems record things the office finds out days later, or not at all — a truck that ran with nobody logged into it, a shift starting with no work on it. Connect them and those become a short list of things to deal with, rather than something nobody looked for.
From the ELD
Your driver and vehicle roster, and the live movement feed, read every five minutes. Retired units are recognised as retired rather than quietly disappearing — on one real account that was 187 of 382 rows.
From the TMS
Shifts, orders, locations, depots and equipment, mirrored locally every fifteen minutes so the board does not depend on your dispatch system being reachable at the moment somebody opens it.
Matched to your records, with your confirmation
Nothing is adopted silently. You see what the vendor reports beside what NAVYAX holds, and you decide what pairs with what — including depots, which is what makes an expiry date get judged in the right time zone.
A connection that can say it is broken
Nine health states, and three of them stop the system retrying until somebody stores a working credential. A feed that stops is visible on the screen rather than discovered a month later.
GhostRun
A truck is moving. Nobody is logged into it.
A unit is moving and the ELD says nobody is logged into it. The hours go against no driver, and somebody has to work out who it was — usually long after anyone remembers. GhostRun watches for it and tells you how long it has been true.
An episode, not an alert
It opens the first time a unit is seen running unattributed, stays open while that is true, and closes itself the moment a driver logs in or the truck stops. What you get is a duration: this unit ran unattributed for fifty-three minutes, and now has a driver.
It decides on the ELD's own driving flag, not on speed
Measured on a live account: three devices reported not driving at speeds up to 90, and speed is an integer whose median is zero. A rule written on speed raises on yard noise and misses real movement.
A login that cannot see drivers raises nothing
If no unit in the whole fleet reports an attributed driver, that is the shape of an ELD login missing driver permissions — not a fleet nobody is driving. The read is distrusted and the board says so, rather than flagging every truck you own.
A stopped feed does not read as all clear
Durations are measured to the last reading, never to the clock, so a dead poller never renders as a worsening incident. The board tells "nothing is wrong" apart from "we stopped being able to look", and says which.
The driver is named, and never told
Names come from your own roster rather than a vendor id. And no driver is ever shown this board — the system refuses them outright. Finding out a truck ran unattributed is the office's job, and the product does not accuse anybody.
What it does not do yet
It watches and it reports. There is no sign-off or dismissal yet, and no per-vehicle exemption for a yard truck that shunts all day — both are on the list, and you would notice their absence in a week.
Shift readiness
Before a shift starts, has the driver got work assigned and a unit ready? Raised ahead of the shift by however long you set, per rule.
The next delivery
When a driver finishes an order, check the details of the next one — the site, the order, the product. On or off per customer, because some need confirming every time and some never do.
A task you could not finish stays open
Somebody who tried and got no answer logs the attempt, and the task stays where it is. That is the difference between a system people use and one they close to make the number go down.
It knows when the work went away
A withdrawn shift closes its check as cancelled rather than overdue, and an order the TMS marks complete drops its check. A poller outage is told apart from work that genuinely vanished.
Sentinel
The checks somebody was supposed to remember
The same idea on the dispatch side. Two checks, raised from what the TMS reports, that exist because the alternative is somebody remembering to look.
How this is switched on
Neither module comes with the product. We turn it on for your account, and then your own people enter the credentials — NAVYAX support cannot do it for you, and the server refuses the attempt rather than relying on us to be careful. If your ELD or TMS is not one of the two we connect to today, we will say so on the call rather than after.
Your credentials are yours
When you connect a system, your own staff enter the credential. NAVYAX support cannot do it on your behalf — not as a matter of policy, but because the server refuses the request and names the reason. The same rule covers confirming an import.
And it reads. Nothing is written back into your ELD or your TMS, so connecting NAVYAX cannot change a record in the system your dispatchers are working in.
Book a demo
Twenty minutes, on your own data if you want
Send a roster beforehand and we will import it first, so what you are looking at is your fleet rather than a made-up one.
You will get a reply from a person, usually the same day. There is no sequence, no sales team and no follow-up drip — there is one of us.
If you would rather not use the form, hello@navyax.com and (227) 268-4898 both reach the same place.