How does a manual SOS differ from automatic inactivity detection?
SOS is active and unconditional. The worker raises it personally when they see a hazard or feel unwell, and the alarm does not depend on inactivity settings or on location configuration. Inactivity is passive: the system starts it only after there has been no movement for a set time.
The second difference concerns closing. The inactivity alarm closes by itself when the worker moves again, and the alarm permit created by the inactivity closes with it. SOS does not close automatically when movement resumes and stays open until a person closes it. A worker who has called for help and then starts walking still has an active call.
In practice, SOS is the tool for a situation in which the worker is able to act, and inactivity is the tool for a situation in which they are not. Both can coexist on one device and in one crew.
How is the watch SOS raised, and can it be cancelled?
The SOS alarm on the watch is raised with a triple press or a long press of the button. After activation a three-second countdown starts, during which the alarm can be cancelled. The window protects against an accidental press with gloves on or during physical work.
Once the countdown ends, SOS creates its own permit and starts the emergency response chain: a messenger notification and automatic phone calls to the people responsible. The alarm stays open until a person closes it. Closing is a deliberate decision by the dispatcher or coordinator, not the result of time passing.
Example warning pattern on an LTE watch
In a sample pattern for an LTE watch, 20 seconds without movement triggers a warning, and a further 30 seconds without a response triggers the alarm. The SOS button works independently of this and raises the alarm and the phone call immediately. The thresholds are a starting point that the team tunes to the nature of the work.
Sample warning pattern on an LTE watch
How does the system detect inactivity, and how do you know it is not a false alarm?
Inactivity detection works on two sides. The mobile application analyses the device sensors, and the server measures time independently. The full alarm starts only when both measurements agree, which reduces false alarms caused by a faulty sensor or a connectivity problem.
Response times, confirmation times and inactivity sensitivity thresholds are configured in the panel. A crew supervisor does not work on factory settings: the threshold for a worker doing a patrol can differ from the one for a person carrying out work inside a chamber.
A working inactivity deployment provides figures that illustrate the practice well: over 3,400 hours of monitoring, a median alarm closure time of 52 seconds, and 7 in 10 stops cancelled by the worker themselves. Most warnings therefore end with a short human reaction, not with a call for outside help.
What happens once the inactivity threshold is exceeded?
After the threshold is exceeded, the system sends a push notification and an e-mail to the people responsible, creates an alarm permit from the ALARM template, marks the alarm on the map and can trigger an external webhook. The dispatcher sees the worker's position on the facility map, not only a message that someone is not responding.
When the worker moves again, the alarm permit closes by itself. This means the event does not need manual cleaning up after a false alarm, and the trace stays in the register with the time and the author of the change.
If the client's network blocks all outbound traffic from the application host, the alarm webhooks are stored durably and an external poller collects them over an inbound channel. The alarm is not lost because of network restrictions.
Stops cancelled by the worker
Will a watch without Wi-Fi still raise an alarm?
Pulses from active LoRa cards feed the same inactivity mechanism as the data from the application. A LoRa bridge translates the frames into the existing API, so the watch does not need Wi-Fi. This matters at distributed facilities, where coverage is often uneven.
A variant with a repeater gateway, which forwards frames to a gateway with internet access, is documented as a decision to be made when a real facility lacks full Wi-Fi coverage. Today we describe it as an option on request, not as a standard.
When should a crew use SOS, and when should it rely on inactivity?
The rule is simple: SOS when the person can and wants to call for help, inactivity as a safety net when they cannot. Both functions should be switched on together for every person working alone.
During training it is worth rehearsing three scenarios: a deliberate call for help, cancelling an accidental SOS within the three-second window, and responding to an inactivity warning. A worker who has seen a warning once and cleared it by moving knows what to expect.
Checklist
Video
Frequently asked questions
Does SOS close by itself when the worker starts moving?
No. SOS stays open until a person closes it, because it is a deliberate alarm. Only the alarm permit created by inactivity closes automatically.
Does SOS depend on inactivity settings or location?
No. SOS is active and unconditional, so it works regardless of the inactivity configuration and location. The map position is an addition, not a condition for the alarm.
Who receives the notification when an inactivity alarm is raised?
The people responsible, as set in the configuration, receive a push notification and an e-mail, and the alarm is marked on the map. The system can also trigger an external webhook.
Can the inactivity threshold be changed without a software update?
Yes. Response times, confirmation times and sensitivity thresholds are configured in the panel, so tuning does not require the vendor to intervene.
Match the alarm mechanisms to your facilities
Book a call with InnerWeb to set inactivity thresholds and the SOS chain for crews working alone at water and wastewater facilities.
Let's talk