Manual data extraction remains one of the hidden bottlenecks in engineering workflows. In the context of SAG mill liner wear modelling, it slows down calibration, introduces inconsistencies, and limits scalability.
During his 6-week Summer Research Program at The University of Queensland, third-year Mechatronics student Rahil Patel supervised by Dr Christian Zualaga Bedoya, worked at Julius Kruttschnitt Mineral Research Centre (JKMRC), he developed an agentic AI-driven workflow to automate key steps in JK-MillFIT calibration.
SAG mills are highly energy-intensive assets, and accurate liner wear information is critical for maintaining performance and supporting optimisation strategies. However, this data is typically buried in wear scan reports and requires significant manual effort to extract, clean, and structure before it can be used.
Rahil’s solution replaces this manual process with an integrated, agent-based system. The workflow interacts with researchers, validates data stored in cloud environments, extracts relevant information directly from report files, and structures it into standardised formats. It then connects seamlessly to analytical tools to automate the calibration of liner wear models used in JK-MillFIT.
The result is not just incremental improvement, it is a shift from a fragmented, manual workflow to a structured and automated pipeline. This reduces reliance on manual data handling, improves consistency and traceability, and creates a scalable foundation for future deployment across operations.
Beyond the immediate application, this work highlights a broader direction: agentic AI systems can play a critical role in transforming how engineering tasks are executed, moving from tool-based support to orchestrated, end-to-end workflows.
At JKMRC, initiatives like this are part of a wider effort to integrate advanced digital capabilities into mineral processing, while providing students with opportunities to work on real industry challenges.
Learn more about the UQ Summer Research Program: https://employability.uq.edu.au/summer-winter-research