Industrial IoT & Intelligence
Turn installed industrial equipment into a system you can observe and maintain
Equipment in the field is easy to install and hard to keep in view. Indeema engineers the path from a machine’s condition to a device, a connection, a comparison against reference levels and the people who act on it, and the means to configure and update the installed devices afterward.
See what changes along the lineWhere this begins
Most equipment monitoring work begins with one of these
Common starting points, not claims about a past delivery.
The equipment is installed, and we cannot see how it is doing from here.
READ · CARRYWe learn that something is wrong when a person on site notices, or when it stops.
COMPARE · ATTENDThe pilot worked on the bench. Turning it into a system we can maintain is the hard part.
CARRY · KEEPEvery device in the field has to be configured and updated by hand.
KEEPReadings are taken by hand, location by location.
READ · CARRY
The asset line
Five changes, from the machine to the installed base
Each stage changes something in your system. The equipment and the controllers that run it stay as they are.
READ
The machine’s condition becomes a signal: vibration, power draw or state.
Hand over Hardware & ElectronicsCARRY
The reading leaves the machine over a link that suits the site.
Hand over Embedded & FirmwareCOMPARE
The reading is judged against a reference level or threshold, so attention goes where it is needed.
ATTEND
A person is told, and acts. The system does not act on the machine.
KEEP
Installed devices are provisioned, configured and updated remotely.
Understanding, in proportion
Compare first. Interpret only when the decision needs it
A reference level or a threshold is often the right answer. Adaptive thresholds are algorithms, not machine learning, and not every equipment estate needs a model.
Show
Put the reading in front of a person.
Compare
Judge it against a reference level, or a threshold adapted to the equipment.
Interpret
Use a model only when the decision needs one and the method can be shown to work on the target.
AI for IoT & Edge AI
Intelligence here means proportion: as much understanding as the decision needs.
Why Cognitive IoTBeside what already runs
The layer sits beside your controllers. It does not replace them
The controllers and platforms that already run the equipment stay in charge. This layer reads, carries and reports, and a person acts. It does not replace or program industrial automation. Three questions settle the boundary before anything is installed.
- What stays in charge of the machine?
- What may the new layer read?
- Who is told, and how?
After installation
Devices in the field stay configured and up to date
Installed devices outlive the project that placed them. Each needs its own credentials, its site’s settings and a way to receive new software. The equipment stays where it is; what changes is how each device is set up, watched and kept current.
- ProvisionEach device receives its own credentials when it first starts.
- ConfigureSite settings are set through one tool, not by hand on each device.
- UpdateNew software reaches devices in the field remotely, and each device’s state stays visible.
The device side is Embedded & Firmware . The service side is Cloud & Connected Systems .
What this has looked like
Three situations, and what Indeema owned in each
Client names and figures are not shown. Each example is a draft case, and each carries its own boundary.
- READ · COMPARE · ATTENDMachinery condition monitoring
A vibration-diagnostics product line for rotating machinery in manufacturing.
- Indeema owned
- The sensor unit and its firmware, the dashboard and the cloud behind it.
- What changed
- Vibration was analyzed against reference levels, with alarms, reports and a diagnostic dashboard for operators.
Boundary Signal analysis and alerts only. Machine learning and scale are not claimed.
View the draft case - CARRY · COMPARE · KEEPEquipment energy and state
A consultancy wanted energy, trip and out-of-service data from existing lifts that start at different currents.
- Indeema owned
- The device beside the meter, its firmware, the cloud connection and portal, then remote updates.
- What changed
- The device reads the meter over RS-485 and Modbus, reports over MQTT with TLS, and adapts its thresholds to each lift.
Boundary Adaptive thresholds are algorithmic, not machine learning. No results are claimed.
View the draft case - CARRY · ATTEND · KEEPField-unit monitoring and provisioning
A platform company for oil, water and gas SCADA units needed devices set up and monitored through one web application.
- Indeema owned
- Gateway provisioning, the web application and dashboards, remote software updates and testing.
- What changed
- One tool for installation and configuration, node status, and remote updates to devices in the field.
Boundary Monitoring and provisioning around the client’s platform. No control of equipment and no results are claimed.
View the draft case
Where to start
Start with what the equipment should report
Product Discovery & Technical Strategy
Work out what your equipment should report before anything is built.
IoT Product Engineering
Have one team own the whole system, from device to service.
Bring us
- The equipment and where it sits
- The signal you want read
- Who should be told, and what they do
Sector context: Manufacturing · Oil & gas