The University of Western Australia (UWA) has worked in partnership with Curtin University, CSIRO, Alcoa, BHP and Roy Hill to improve the maintenance duties of large processing plants.
The research was performed as part of the ARC Training Centre for Transforming Maintenance through Data Science and sought to dive deep into how, when and why maintenance occurs in the resources sector.
This was done using ‘network science’, involving computational methods, statistics, applied mathematics and artificial intelligence, according to the UWA.
UWA School of Physics, Maths and Computing professor Michael Small said the project was inspired by a realisation that maintenance was commonly followed-up by corrections to the plant.
“It’s a problem known as a cascading failure, but identifying and confirming these events is incredibly difficult when you take into account the number of assets and the volume of maintenance and operating data we’re talking about in companies of this size,” Small said.
“Our industry partners run large and complicated plants full of equipment and they have mountains of records stored in systems like SAP that capture each and every time maintenance is carried out.”
The research team worked with a mining company to understand maintenance to its pumps, involving 88,545 work orders for 5655 different pumps over eight years.
Aside from using mathematics and modelling, the project consulted with engineers who could report on the ins and outs of plant upkeep, according to Small.
“But it wasn’t just the work orders, we worked with the maintenance engineers on the ground who have the spanners in their hands and who often have that unwritten, institutional knowledge, to model the behaviour of the past to try and find patterns using complex network analysis,” Small said.
“Using our model, which has some very heavy mathematics behind it, we were able to identify pumps that were ‘super spreaders’ – pumps which experienced corrective maintenance events which lead to corrective maintenance on other pumps.”
Small said the results were intriguing for researchers and engineers alike.
“Interestingly, we found that there were ‘self-loops’ in certain pumps, where an asset fails and is repaired and then fails again, and also that there was a prevalence of hidden failures in standby pumps,” he said.
The model continues to be developed for accuracy and heightened understanding of where resources companies can improve their processes to remove failures in maintenance.
“It’s been rewarding doing something positive and applying data science to something that will ultimately help companies make real cost savings,” Small said.