MACSI has won the Knowledge Transfer Ireland Consultancy Impact award, for a project with Rusal Aughinish Alumina, run by William Lee and Vincent Cregan.
The Rusal Aughinish Alumina (AAL) plant in Limerick extracts alumina from bauxite using the Bayer Process and is one of the world’s most energy efficient alumina refineries globally. AAL has a long-standing relationship with the Mathematics Applications Consortium for Science and Industry (MACSI) at UL with whom they have successfully collaborated for over 10 years on a number of research projects. AAL sought to improve predictions of product quality five days in advance to address the fact that system events meant the plant could lose power directly impacting product quality. AAL engaged the MACSI team on a piece of consultancy tasking MACSI with improving the prediction process. To do this MACSI developed mathematical and statistical models for process understanding that combine first principles simulation and data driven stochastic dynamics. The project resulted in a 200% increase in the accuracy of prediction of product quality, enhancing the overall efficiency of the plant and supporting AAL decision making. The two algorithms developed under the consultancy engagement have been implemented by the company and are in daily use. This project has had a major impact on the MACSI team at UL whose skills have been enriched by applying mathematical techniques to real world industrial problems. It enabled two-way knowledge transfer between the parties through staff exchanges and joint workshops.
AAL, MACSI and UL continue to work together with this project having contributed to the development of a new SFI-funded research agreement in 2016 and a new project between AAL, UL and Imperial College London