Schneider Electric AI Vision Inspection Boosts Throughput 70x for Automotive Parts

Title

Schneider Electric AI Vision Inspection Boosts Throughput 70x for Automotive Parts

Content

Schneider Electric, working with machine vision partner Cognex for over 20 years and joining Schneider's Technology Partner Program in 2018, has onboarded nearly 100 smart factories since launching its Smart Factory Program in 2017 and reduced energy costs at individual plants by 10 to 30 percent and maintenance costs by 30 to 50 percent by integrating Cognex machine vision, deep-learning inspection, and barcode-reading technology with the EcoStruxure IIoT platform across manufacturing, logistics, and warehousing operations. According to Laurent Chantoiseau, GSC Europe Data and Analytics Director at Schneider Electric, "Cognex has been a key partner for Schneider Electric as we have taken our digital transformation journey. Cognex solutions are certified and part of our standards for deployment on our production lines. Our relationship with Cognex is based on a strong partnership," while on ease of deployment he added, "Cognex edge learning solutions are incredibly easy to set up; generally, in less than 4 hours for a simple use case, we can have an application ready for production," and on long-term direction he noted, "As a partner, we rely on Cognex for our projects. Even though we have succeeded in working on a wide range of improvements with deep learning, we wish to go even further with our inspections to protect our customers. We will always work together to improve our supply chain." Schneider standardized globally on Cognex machine vision and barcode reading technology for four reasons: high accuracy in detecting product flaws, fast and easy integration via standard communication and networking protocols, fast setup with set-up wizards, and versatility that makes it easy to learn and replicate at both plant level and global scale. At the Chasseneuil, France site, Cognex edge-learning classification tools identify the presence of contact pads and correct spring assembly in Tesys GS modular contactors, replacing earlier rule-based vision systems that produced inconsistent quality and unacceptable false-reject rates, while a separate VisionPro Deep Learning application automates multi-part cabinet assembly inspection, locating mounting rails and electronic modules within each cabinet and verifying part numbers against the assembly plan, both feeding results back into production parameters to mitigate process drift and improve end-to-end efficiency. Schneider plans to continue the partnership with more machine-vision-based quality control across its manufacturing operations and distribution centers, with priority areas being earlier in-process quality checks, reducing false rejects, and improving sustainability metrics including scrap, raw material consumption, and energy use.