South KoreaManufacturing2021 – ongoingDoosan Enerbility
D-Vision AI nondestructive inspection
Doosan Enerbility reported 3 years of factory use for its D-Vision inspection AI
Problem
Equipment inspection depended on specialists interpreting images by eye. Factories faced shortages of qualified inspectors while safety requirements demanded reliable detection.
Approach
Doosan Enerbility developed D-Vision to interpret nondestructive inspection images. It trained the model on more than 100,000 images and applied it in manufacturing for 3 years by late 2024. An executive reported shorter inspection work and improved accuracy without publishing numerical improvements.
Results
No published figures
Takeaway
The company reported operational benefits, but did not quantify changes in inspection accuracy or cost.
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
ZDNet reported an executive's account of D-Vision's development and factory deployment. The company said the system reduced inspection time and cost while improving accuracy.
Full details are in the original sources.