At Eramet, we see Digital Transformation as a lever to create safer, more efficient, and more sustainable mining and metallurgical operations. By combining advanced analytics, AI, automation, and domain expertise, we look forward to building the next generation of industrial performance.

Our focus spans breakthrough capabilities such as Holistic Predictive Maintenance, AI‑based furnace process optimization, and autonomous or robotic systems that remove people from hazardous environments. These smart solutions help us reduce downtime, improve asset reliability, enhance energy efficiency, and create more resilient production systems.

Digital innovation is not only about technology — it’s about empowering our teams with actionable insights and transforming the way we work, operate, and collaborate across the value chain.

AI-based furnace process optimization

Furnaces play a critical role in mineral processing, and their performance directly impacts energy consumption, throughput, and product quality. Eramet operates both open and closed furnaces in its refining plants. Thanks to the availability of operational data, our metallurgists have successfully categorized the state of our processes, providing a strong foundation for AI-based solutions.

Eramet has been involved from the early stages in exploring the use of Artificial Intelligence (AI) to make furnace operations smarter. Our internally developed algorithms recommend adjustments to maintain stable process conditions and improve operational efficiency.

With the advent of new AI models and tools, we are now looking to scale and enhance these efforts by exploring cost-effective, scalable, and adaptive AI solutions that can support real-time monitoring, prediction, and optimization of furnace processes across different sites and furnace types.

Key Interest Areas

  • Support metallurgists and operators with actionable insights and decision-support tools.
  • Predict anomalies or failures before they occur, enabling proactive maintenance and reducing downtime.
  • Recommend process adjustments in real time to optimize energy use, throughput, and product quality.
  • Solution should be scalable and preferably at TRL 8 and above.

Solutions not of Interest

  • Focus solely on basic data visualization without predictive or prescriptive capabilities.
  • Require complete replacement of existing infrastructure or control systems.
  • Are generic AI platforms without proven relevance to furnace operations or mineral processing.
  • Models without transparency or explainability for process engineers.
  • Lack of a cost-effective scalable model.
  • Solutions that are purely theoretical and have not yet been validated at small scale or demonstrated in any adjacent application.

Note: Please include a non-confidential explanation of your solution when submitting.