Detect equipment degradation before failure

Continuous health monitoring of HVAC and critical systems. Anomaly detection models provide 7-day advance warning, reducing unplanned downtime by 45%.

Predictive maintenance

Move from reactive firefighting to planned precision

When an AHU fails in a hospital, the impact goes beyond the equipment itself: operating rooms lose classification, cleanrooms breach compliance, patient comfort degrades. Emergency repairs cost 3-5x more than planned maintenance.

Calendar-based maintenance follows manufacturer recommendations — but actual equipment wear depends on operating conditions, load patterns, and environmental factors. The result: some components are serviced too early, while others fail between intervals.

-45%

Unplanned downtime thanks to early anomaly detection

7 days

Average advance warning before predicted failure

3-5x

Emergency repair cost premium vs. planned maintenance

TERA 360 dashboard displaying equipment health scores and degradation trends

The challenge

Unplanned failures cost far more than just repairs

Bearing wear, belt slippage, refrigerant leak, control valve drift — these failures develop over weeks or months. Without continuous monitoring, degradation is invisible until it causes a functional failure or a comfort complaint. By then, secondary damage has often occurred.

In pharmaceutical production, unplanned HVAC failure can destroy entire batches worth hundreds of thousands of euros.

Facility maintenance teams spend most of their time responding to breakdowns and complaints rather than preventing them. Spare parts are either overstocked (capital tied up) or understocked (delays in repair). Work planning is disrupted by emergencies.

How TERA 360 works

From sensor data to maintenance intelligence

TERA 360 monitors equipment health continuously and applies anomaly detection models to identify degradation patterns before they cause failure.

Continuous health monitoring

Vibration sensors on rotating equipment, DeltaP trends across filters and coils, energy signatures from motors and compressors, runtime hours and cycle counts. Every data point contributes to an equipment health profile that evolves over time.

Anomaly detection

Machine learning models trained on normal operating patterns detect deviations: increasing vibration amplitude, abnormal energy consumption, DeltaP drift beyond loading curves, unusual cycling frequency. Each anomaly is scored by severity and rate of change.

Predictive scheduling

Detected anomalies are translated into maintenance recommendations with estimated time to failure, required parts, and labor estimates. Recommendations are grouped to optimize maintenance windows — minimizing disruption while addressing all identified issues.

Sensors & systems

Equipment health instrumentation

Vibration sensors

Tri-axial accelerometers on fan bearings, pump housings and compressor mounts. Frequency-domain analysis detects imbalance, misalignment, bearing wear and looseness patterns.

Electrical monitoring

Motor current signature analysis on fans, pumps and compressors. Power quality metrics detect winding degradation, phase imbalance, and VFD faults. CT-based, non-invasive installation.

Process parameters

DeltaP across coils and filters, supply/return temperatures, refrigerant pressures and superheat. These process parameters reveal degradation that vibration or electrical analysis alone would miss — fouled coils, refrigerant loss, stuck valves.

Runtime & cycle tracking

Equipment operating hours, start/stop cycles, and mode transitions. Excessive cycling indicates control issues. High runtime hours trigger maintenance milestones. Standby equipment is monitored for readiness.

TERA 360 dashboard with equipment health scorecard and anomaly timeline

Dashboards & alerts

Maintenance intelligence for facility teams

Unified view of health status across all your monitored equipment, with drill-down to individual sensor data and anomaly history.

Health scorecard — Every asset displays a composite score (0-100) color-coded from green (healthy) to red (act now). Historical trends reveal degradation rates.
Anomaly timeline — Chronological view showing affected equipment, anomaly type, severity score, estimated time to failure, and recommended action. Pattern analysis identifies systemic issues.
Maintenance planner — Actions grouped by urgency, location and trade. Auto-generated parts list with stock levels and supplier lead times. CMMS integration for automatic work order creation.
Performance analytics — Track KPIs: percentage of unplanned vs. planned work, mean time between failures, prediction accuracy, cost avoidance. Monthly reports demonstrate ROI to management.

Benefits

From reactive firefighting to planned precision

-45%

Unplanned downtime. Anomaly detection catches degradation weeks before failure. Maintenance is scheduled during planned windows, not triggered by emergency calls.

-30%

Maintenance costs. Elimination of emergency repair premiums, reduced secondary damage, optimized parts inventory, and fewer unnecessary scheduled interventions.

Extended equipment life, early warnings

Early detection of wear patterns allows corrective action before damage propagates. Bearing replacement before failure prevents shaft damage. Refrigerant leak repair before compressor failure. Each intervention extends the useful life of the equipment by an average of 20%.

Average lead time from anomaly detection to predicted failure: 7 days. Sufficient time to order parts, schedule labor, coordinate with operations, and plan the maintenance window without disrupting critical activities. Critical environments maintain compliance continuity.

Maintenance labor is redirected from reactive to proactive work.

45%

Less unplanned downtime

30%

Maintenance cost reduction

20%

Extended equipment life