Tafalla Iron Foundry (TIF), an industrial cooperative, announced that it has successfully completed the development phase of its R&D project “Optimodel,” aimed at optimizing the operation of its melting furnace through the use of advanced predictive models. Specifically, the initiative has enabled the company to reduce defects associated with metallurgical causes by up to 50% and improve process stability, strengthening data-driven decision-making in one of the most critical stages of the foundry process.
The “Optimodel” project, launched in 2024 and scheduled for completion in 2026, focuses on monitoring and optimizing the iron melting process. To achieve this, Tafalla Iron Foundry has integrated industrial sensors, process data acquisition and management systems, and machine learning algorithms, with the goal of anticipating the evolution of the molten metal composition and optimally adjusting furnace charge recipes.
Thanks to this approach, the cooperative has advanced process control and optimized the consumption of critical and alloying materials, without compromising the technical specifications of the final parts. Results indicate reductions of up to 15% in copper consumption and up to 50% in tin usage, along with additional decreases in other ferroalloys, contributing to greater efficiency and sustainability in the production process.
Another key milestone achieved during this phase has been the development and technical validation of predictive models capable of estimating the evolution of the molten metal composition based on the input charges and furnace operating variables. These models make it possible to anticipate key material properties, such as mechanical strength or hardness, before the parts are manufactured, reducing uncertainty and the need for subsequent corrections.
“We have moved from an approach based on corrective controls to a predictive model that allows us to anticipate and make more accurate decisions,” explains José Antonio Molina, Industrial Director of Tafalla Iron Foundry. “The application of advanced analytics to the melting process has a direct impact on product quality and places us in a very strong position to move into the next phase of industrial implementation,” he adds.
“Optimodel” is being developed in close collaboration with the Public University of Navarre (UPNA), which has contributed its expertise in advanced data analysis and mathematical modeling, representing a clear example of university–industry knowledge transfer applied to a real industrial environment. The project is also co-funded by the Government of Navarre, as part of its policy to support industrial R&D and the digitalization of the region’s productive sector.
Having completed the development phase, Tafalla Iron Foundry is now preparing to begin its industrial deployment stage, which will include the progressive integration of predictive models into plant control systems and the execution of pilot tests under real production conditions throughout 2026.
