Models require continuous attention
Equipment changes, new operating conditions, or outdated training data can reduce model relevance.
Protect availability with alerts teams can trust
Detect early signs of equipment degradation, qualify alerts with operating context, and coordinate action.
6000+monitoring projects in service
The operational reality of APR
Early detection creates value when teams can act on the alert
An APR* alert starts an investigation. Before addressing it, teams need to assess its relevance and understand it.
*APR: Advanced Pattern Recognition
- Operational change
Noise erodes confidence
Alerts should filter routine noise so teams can focus on meaningful signals. Loose thresholds, unsuitable filters, or sensor issues can erode alert quality and consume team time.
Sensor driftAlerts need context
Teams need to understand what changed and under which conditions before deciding what to do next.
- Operating mode
- Unknown
- Contributing signals
- Not ranked
- Maintenance context
- Scattered
Why Metroscope
Build anomaly detection around plant reliability priorities
Maintain detection quality as the plant changes
Metroscope helps teams build and maintain reliable models through representative training data and alert configuration. It flags new operating conditions outside the models’ training data and offers quick retraining to keep monitoring up to date.
Sensor Health Management
Reliable models depend on reliable sensor data
Automatically detect degraded or missing sensors, identify affected models and alerts, manage exclusions, and track repairs.
Evidence from plant applications
Discover real plant challenges with results
Reactor coolant pump monitoring
A dedicated strategy for reactor coolant pump long-term operation
Metroscope translated failure modes and available instrumentation into 13 monitoring models. Historical replay identified lower-bearing drift six months before the known outage.
Biomass plant value estimate
€180k in potential annual value for a biomass plant
For a 25.2 MW net biomass plant, the business case estimates potential annual value across performance, maintenance, and availability.
Delivery and support
Adapt anomaly detection to your monitoring organization
Use Metroscope with your own experts, or extend your team with support for model management and alert qualification.
Direct operation
Operate the software with your own team
For organizations with the monitoring expertise to manage models and alert workflows directly.
- Your team
- Runs daily monitoring, manages alert rules, and reviews model performance.
- Metroscope
- Supports scope definition, data preparation, setup, validation, and training.
Supported operation
Add ongoing monitoring support
For teams that want support with model maintenance and first-level alert qualification.
- Your team
- Provides operating context, confirms findings on site, and coordinates maintenance.
- Metroscope or an approved partner
- Reviews and pre-qualifies alerts, maintains models, and adjusts detection settings as conditions change.
A 6-week path to launch
Choose the availability risks to cover
Define scope, data, and responsibilities
Configure, validate, and launch
Train your teams and ensure adoption
Define the monitoring and support that fit your plant