Fix the machine before it breaks.
What It Is
Most failures announce themselves. A bearing starts to vibrate at a new frequency, a motor draws a little more current, a gearbox runs a few degrees warmer. Predictive maintenance captures those signals continuously and compares them with how the asset normally runs.
Simple rules catch the obvious cases; machine-learning models catch the slow drifts and the combinations a person would miss. The output is a ranked list of assets that need attention, with the evidence, sent to the maintenance team and their CMMS.
How It Works
The system from the sensor to the people who act on it. Every project is sized to the site, but the layers stay the same.
- Sense
Condition sensors
Accelerometers, current transformers, temperature and pressure transmitters on the assets that matter most.
- Edge
Edge gateway
Samples high-rate vibration, runs FFT and envelope analysis, and reads PLC tags for speed and load.
- Connect
Plant network
Features and alarms published over MQTT or OPC UA through a segmented OT network.
- Store
Historian
A time-series historian keeps the trends that models are trained and re-trained on.
- Act
Models & alerts
Anomaly scores and health indices on a dashboard, with work orders raised in the CMMS.
Key Technologies
The protocols, standards and components we work with on this use case.
Sensing
- IEPE / MEMS accelerometers
- Motor current signature analysis
- RTD / thermocouple
- Ultrasound
- Oil & particle sensors
Analysis
- FFT spectra
- Envelope demodulation
- Anomaly detection
- Remaining useful life
- ISO 20816 vibration severity
- ISO 13374
Connectivity
- OPC UA
- MQTT
- Sparkplug B
- Modbus TCP
- PROFINET
- EtherNet/IP
Systems
- Time-series historian
- CMMS integration
- SAP PM / Maximo
- Grafana-style dashboards
Typical Deliverables
What a project usually hands over. We agree the exact scope with you before work starts.
- Criticality review and a list of the assets worth instrumenting
- Sensor selection, mounting and wiring drawings
- Edge gateways configured and commissioned on site
- Historian, dashboards and alarm limits per asset
- Trained anomaly models with a plan to re-train them
- Hand-over to the maintenance team, with CMMS work-order integration
Where It Applies
Industries where this use case answers a problem we see often. Each links back to the sector page.
Related Use Cases
Edge AI
AI models that run next to the machine, fast and without depending on the cloud.
Learn morePLC, SCADA & historians
Controls engineering and a live SCADA that ties the floor together.
Learn moreRemote asset monitoring
IoT sensors and edge devices that watch distant or scattered assets around the clock.
Learn more
Let's Scope Your Project.
Tell us about your site and what you want to improve. Our engineers will help you scope a practical first step.