Your machine will fail.
We're building the system that tells you why.
Achieve end-to-end predictive maintenance for your critical industrial assets with CauseNex. Through a precise mix of plug-and-play hardware and proprietary Causal AI, we equip maintenance teams with actionable, definitive root-cause diagnostics. Protect your motors, gearboxes, pumps, and CNC systems, and eliminate unplanned downtime entirely.
The Problem
Bearing failures cost German factories millions every year — without warning.
In most manufacturing facilities, rotating equipment fails without predictable warning. When a motor or pump bearing fails unexpectedly, the costs include lost production, emergency repairs, unplanned overtime, and expedited parts orders.
73%
of industrial bearing failures give no detectable signal before breakdown
Source: Industry estimate
€18K
estimated average cost per unplanned failure in German SME factories
Source: Fraunhofer IFAM Bremen
9 hrs
average unplanned downtime per failure event
Source: German manufacturing industry data
Our Approach
We are building sensor probes that explain why — not just that.
Current condition monitoring systems detect anomalies. They tell you something is wrong. CauseNex is designed to go further: our approach uses causal AI methodology to reconstruct the physical mechanism behind each fault.
The difference is significant. An anomaly alert says:
"Bearing fault detected. Confidence: 94%."
A causal explanation says:
"Shaft misalignment is causing resonance in the bearing housing, accelerating wear on the outer race. Estimated time to failure: within the next 14 days."
The maintenance engineer can verify the causal chain physically. They can act on it. They can explain it to management. This is what we are building.
× Standard anomaly detection
BEARING FAULT DETECTED Confidence: 0.94 ...and that's it.
✓ CauseNex approach (in development)
INNER RACE FAULT — DEVELOPING Root cause chain: Shaft misalignment → bearing resonance → outer race wear → estimated failure: ~14 days Each step physically verifiable.
This approach is based on M.Sc. research at Constructor University Bremen. The product is currently in development. No commercial product is available for purchase at this time.
Ecosystem
Developed with support from academic and public innovation ecosystems.
Constructor University Bremen · Constructor Start