RUSAL to save over $10 million with AI-based system for electrolytic cells

RUSAL to save over $10 million with AI-based system for electrolytic cells

RUSAL is to implement an AI-based predictive system developed by Mechanica AI, a provider of AI-based solutions for process industries. The expected annual yield growth following the system roll-out is estimated to be over $10 million yearly for the first chosen plant.

RUSAL is one of the world's largest aluminium producers, and ongoing optimisation of production processes is an important part of the company’s strategy. To investigate whether innovative AI technologies can further support this task, RUSAL successfully completed a pilot project with Mechanica AI. 

Using accumulated historical data, Mechanica AI developed a machine learning system able to predict which electrolytic cells are likely to underperform in the near future. The goal was to address the periodic efficiency decrease of electrolytic cells that leads to lowered production output and thus lost revenue. 

The machine learning system was integrated with existing process databases to receive input data on daily plant operations. For output, it generated a list of cells that were most likely to have an efficiency decrease in the next few days, allowing on-site experts to inspect them as part of their regular maintenance workflow. The system was tested on two RUSAL plants. 

Pilot results demonstrated that the machine learning system correctly identified at least twice as many underperforming cells compared to the current expert-led approach. Early alerts on unnoticed technical problems allowed plant experts to treat them in time, thus avoiding yield loss. 

The predictive system is now being deployed at the first chosen plant. Once the system is fully rolled out, the potential effect for this plant is estimated to be over $10m in yearly revenue. The gains will result from the timely prevention of cell inefficiencies and a corresponding increase of production levels by up to 0.7%.