Teaching AI to read like a maintenance engineer
If AI systems could read, it could unlock rich archives of organisational knowledge – but how do we teach them?
Supported by MRIWA, PhD student Tyler Bikaun is working at the Centre for Transforming Maintenance through Data Science in Perth to teach computer systems how to read and understand technical reports and engineering requests.
By providing computers with the capacity to understand and extract meaning from written records, Tyler’s work could unlock a valuable resource – years of data on the day-to-day engineering work and maintenance required to keep mining and mineral processing systems running smoothly.
This insight could help deliver value from advanced data science systems in the mineral sector.
By comparing records of operational performance and breakdown with the actions and observations of maintenance engineers, artificial intelligence systems may be able to better predict the maintenance needs of complex mineral processing and mining equipment, supporting better decisions about where engineers should be focusing their work.
MRIWA is partnering with the Australian Research Council (ARC), Curtin University, University of Western Australia, CSIRO, Alcoa, BHP, Roy Hill and the CORE Innovation Hub in the Centre for Transforming Maintenance through Data Science.
The Centre seeks to deliver the next generation of data science solutions focused on the problems that industry needs addressed to deliver efficient and effective maintenance.
Page was last reviewed 8 June 2021