South KoreaEnergy and utilities2024 – ongoingKyungdong City Gas
AI recognition of gas-meter identification numbers in field work
Kyungdong City Gas reduced meter replacement work from 6 steps to 3 with AI number recognition
Problem
Field engineers manually entered gas-meter information into mobile workflows. Meter removal and replacement required a 6-step procedure.
Approach
Kyungdong City Gas applied deep learning and machine learning to recognize meter identification numbers. The system read the numbers and retrieved the corresponding meter information for mobile field work. Company researchers reported its application across 8 work types.
Results
Takeaway
The reported change removed workflow steps through automatic identification and record retrieval. It did not quantify time saved per replacement.
Categories
- Industry
- Energy and utilities
- 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
Kyungdong City Gas researchers described the deployed recognition system in the autumn conference proceedings. They reported a reduction from 6 workflow steps to 3 and application across 8 field-work types.
Full details are in the original sources.