Industrial AI & computer vision
Quality control, predictive maintenance.
Industrial AI is not a research topic: it is quality control that does not tire, maintenance that warns instead of reporting, a setting that adjusts without waiting for the technician's round.
Vision and quality control
Surface defect detection, label checking, presence-absence verification, contactless measurement. The model learns on your parts and your real defects — a generic model does not recognise what you call a defect.
Predictive maintenance
Vibration analysis, electrical signature, temperature, cycle counts. The aim is not to predict failure to the minute but to gain enough lead time to order the part and plan the stop.
Data before model
An industrial AI project spends most of its time on data: historisation, qualification, labelling. We say so before starting, because it determines whether the project will deliver.
Use cases
- In-line appearance and conformity control
- Early process drift detection
- Predictive maintenance on rotating assets
- Setting and scheduling optimisation
- Documentary assistance for operators
Disciplines involved
See the technology in context
We run demonstrations on the Tangier technical floor, on cases close to yours.
Technology as an enabler, never as an end