South KoreaManufacturing2019 – ongoingKG Steel
Image recognition to avoid unnecessary coil-end trimming
KG Steel reported annual savings of KRW 900 million by using AI to avoid unnecessary coil trimming
Problem
KG Steel received steel coils with different end treatments from different suppliers. Cutting the ends from every coil discarded usable material and reduced yield.
Approach
A process-management employee began an AI initiative in 2019 to improve material yield. KG Steel trained an image-recognition system to identify coils whose ends did not need cutting. The employee reported annual savings of KRW 900 million.
Results
Takeaway
The system identified exceptions to a blanket trimming practice. Its reported savings came from retaining usable steel.
Categories
- Size
- Large enterprise
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer vision
- Adoption
- Built in-house
- Goal
- Cost savings
- Evidence
- Company's own claim
Sources
Source summaries
Dong-A Ilbo visited KG Steel's Dangjin plant and interviewed its AI implementation lead. He attributed annual savings of KRW 900 million to image-based selection of coils that needed no end trimming.
Full details are in the original sources.