GermanyManufacturing2025 – ongoingPorsche
Deep-learning paint inspection at Leipzig
Porsche reported scanning and evaluating painted car bodies in under 100 seconds at Leipzig
Problem
Manual paint inspections took time and depended on inspectors spotting small surface defects. Missed flaws required correction before vehicles could leave production.
Approach
Robot-mounted sensors scanned every series-production body shell at Leipzig. Image-processing computers examined the scans, while a deep-learning module classified defects and tracked recurring patterns. Finishing staff received defect locations and close-up images for repairs.
Results
Takeaway
Automated inspection directed finishing staff to specific defects. Employees continued to perform the paint corrections.
Categories
- Size
- Large enterprise
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer vision
- Adoption
- Vendor solution
- Evidence
- Company's own claim
Sources
Source summaries
Porsche described automated paint inspection across series production at Leipzig. It reported about 100,000 images per inspection and a complete scanning and evaluation time below 100 seconds.
Full details are in the original sources.