JapanManufacturing2025 – ongoingSumitomo Metal Mining
AI predictive maintenance at nickel and Harima plants
Sumitomo Metal Mining deployed AI equipment monitoring at 2 Japanese sites
Problem
An unexpected equipment failure could interrupt connected smelting processes. Maintenance decisions depended partly on experienced employees' judgment.
Approach
The company developed machine-learning models with support from NSSOL, using historical abnormality records and experienced workers' knowledge. Sensors supplied equipment measurements, including sound and vibration, for cloud-based anomaly detection. The system entered use at its nickel plant and Harima operation.
Results
Takeaway
Experienced workers' observations became inputs to equipment monitoring. The company did not disclose a measured reduction in downtime or maintenance costs.
Categories
- Size
- Large enterprise
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Partner co-development
- Evidence
- Company's own claim
Sources
Source summaries
DIGITAL X reported the company's announcement of deployment at 2 sites. The company claimed results on more than half the equipment, without defining a quantified downtime benefit.
Full details are in the original sources.