South KoreaRetail and e-commerce2025 – ongoingKurly
Deep-learning inspection of fresh produce
Kurly deployed AI produce inspection at Gimpo and Pyeongtaek with reported accuracy of at least 95%
Problem
Manual visual inspections produced differences in how workers judged incoming fresh produce.
Approach
Kurly introduced camera-based inspection machines at its Gimpo and Pyeongtaek logistics centers in May 2025. A deep-learning sorting solution assessed produce images against 25 criteria. Products scoring below the acceptance threshold were returned.
Results
Takeaway
The machines applied common acceptance criteria across the operating centers. Wider deployment was still planned when the results were reported.
Categories
- Industry
- Retail and e-commerce
- Business function
- Production and qualityLogistics and supply chain
- Innovation type
- Process innovation
- AI technology
- Computer vision
- Adoption
- Vendor solution
- Evidence
- Company's own claim
Sources
Source summaries
Electronic Times reported that Kurly had operated AI produce-inspection machines at Gimpo and Pyeongtaek since May. It described 25 assessment criteria, measurement times of 5–10 seconds and accuracy of at least 95%. Expansion to other centers remained planned.
Full details are in the original sources.
