South KoreaManufacturing2017 – ongoingFrontec
Machine-learning inspection of automotive nuts
Frontec put machine-learning inspection into production to detect small flaws in nuts
Problem
Frontec's existing visual inspection equipment checked nuts against specifications. It missed small surface flaws that customers wanted removed.
Approach
Frontec worked with Kyung Hee University professor Lee Kyung-jeon on a machine-learning inspection system, with support from the Korea Industrial Complex Corporation. Research began in 2017, and the system entered production in 2018.
Results
No published figures
Takeaway
The inspection task extended beyond dimensional compliance to cosmetic defects. Frontec did not disclose a detection rate or financial return.
Categories
- Size
- Small business
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer visionPredictive analytics and ML
- Adoption
- Partner co-development
- Goal
- Quality and risk
- Evidence
- Company's own claim
Sources
Source summaries
The report described Frontec's university partnership and production deployment. It said machine learning detected minor defects that the previous equipment missed, without reporting a quantified operating result.
Full details are in the original sources.