Škoda Auto is employing AI-based image recognition to swiftly identify maintenance needs on the assembly line.
The Škoda FabLab at the main plant in Mladá Boleslav has implemented a system that continuously monitors the assembly line equipment, connected to a computer using Artificial Intelligence (AI) to detect irregularities and necessary maintenance work.
Miroslav Stejskal, Predictive Maintenance Coordinator, Škoda Auto Central Technical Service, says that the company is committed to predictive maintenance as a means to ensure equipment remains in top form. The AI system continually checks the current process status against the optimal conditions, identifying patterns and adding information to its knowledge base, allowing Škoda to react quickly, optimise its efficiency, and make use of production capacities.
Cameras on the overhead conveyor capture images of equipment and parts prone to wear, flagging irregularities or the need for maintenance. “Magic Eye” compares its high-precision photographs against thousands of stored images, identifies differences from optimal conditions and sources of errors. Blue lighting ensures reliable differentiation between cracks and scratches, and the knowledge base expands as worn parts are marked error-free once replaced and rechecked.
The system evaluates detected deviations using past irregularities, and Škoda uses “Magic Eye” at its main plant in Mladá Boleslav on the assembly line for the Enyaq iV and Octavia models, enabling AI predictive maintenance in the Czech Republic. The FabLab has simulated a section of the assembly line to further optimise the system and accelerate wider integration at the Mladá Boleslav and Kvasiny sites, with an “implementation arena” for testing camera settings, system configurations, and simulated damage.
