South KoreaManufacturing2019 – ongoingSamsung Electro-Mechanics
Deep-learning appearance inspection for MLCCs
Samsung Electro-Mechanics deployed deep-learning MLCC inspection and reported better defect detection
Problem
Samsung Electro-Mechanics needed to detect appearance defects in miniature multilayer ceramic capacitors. Existing inspection equipment also left products requiring reinspection.
Approach
Internal AI specialists developed an appearance sorter using existing inspection-equipment data and deep learning. By December 2019, MLCC inspection projects were already deployed at domestic and overseas sites. The company reported better defect-detection accuracy and an improved reinspection rate.
Results
No published figures
Takeaway
The company reported operational inspection improvements without quantifying them. Its wider AI programme's projected savings were separate from realized results for this sorter.
Categories
- Size
- Large enterprise
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer vision
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Yonhap reported Samsung Electro-Mechanics' announcement that AI MLCC inspection had reached domestic and overseas production sites. The company described improved detection and reinspection performance but provided no numerical result for the sorter.
Full details are in the original sources.