On this page (3)
The TraxSystem security terminal works without a signal: the patrol route is downloaded to the device in advance, and everything that happens during the patrol is saved locally. Once back on the network, the terminal sends the records to the server by itself, in batches and without duplicates. The functions that need live data, namely gate handling and random inspections, do require a connection.
Two in the morning, the far end of the storage yard. The Wi-Fi ran out at the weighbridge, the next checkpoint is behind a row of containers, and the phone shows no bars. The patrol still has to be done, and in the morning someone will ask whether it was done in full. In a place like this, an app that needs a constant connection to the server stops being a working tool.
What the terminal records during a patrol
A checkpoint is confirmed by position: the terminal calculates the distance to the checkpoint and saves it with the confirmation, so the record shows not just “scanned” but “scanned from where the checkpoint actually is”. A checkpoint can come with a checklist, instructions and a mandatory photo. Between checkpoints the guard can add a note or a photo, or report an incident straight away with its location, without breaking off the route. The GPS trace is recorded throughout the patrol, including with the screen off, and only during the patrol; outside it the terminal doesn’t record position.
A route can enforce the order of checkpoints or leave it open. An optional checkpoint can be skipped, but the skip stays in the record, so the log shows the patrol as it actually was.
How the terminal sends data once the signal is back
On its main screen the terminal shows how many records are waiting to be sent and whether the server can be reached. When the network comes back, the data goes out by itself, in batches: first the patrol record, then each photo separately, and finally the closure. A dropped connection halfway through deletes nothing, and a retry doesn’t create duplicates, because each record carries its own identifier. An interrupted patrol, even after a flat battery, resumes from its saved state instead of starting again.
The terminal’s other modules work the same way: incidents, warehouse withdrawals outside the warehouse’s working hours, issuing and returning company vehicles with photo documentation, and statements signed at the scene. Statement templates with their translations are stored on the device, so a signature under content in eighteen languages can be collected even with no signal.
The guard doesn’t log in on the terminal. The device is assigned to a post, and the name of the person on duty is taken from the duty log, so every record has an author without anyone typing passwords in gloves.
What the security terminal will not do without a network
Gate handling and random inspections work on live data, so they need a connection to the server: a camera view, a plate read and the list of people selected for inspection have no version from yesterday that would be worth anything.
The second limit lies in GPS itself: satellites need open sky, so a checkpoint in a basement or in the middle of a hall is harder to confirm than one by an outside wall. Take that into account when laying out the route, before the first disputed patrol.
See also: Guard tour checkpoints: patrol records you don’t have to take on trust · Incident documentation is created on the spot or not at all · What is a security duty log