Cameco is a Canadian energy firm that started its digital transformation in earnest in 2019. Its longest-running initiatives are in operations management, where it seeks to improve operational decision-making, safety management and sustainability.
A key element in these initiatives is asset maintenance, representing approximately 25% of Cameco’s overall operating costs at its mining operations. Improving asset management began with automating data collection.
Cameco had struggled to analyze multiple regular condition monitoring data from assets in the past. As a result, the company found it hard to understand what went wrong in the event of asset failure.
Senseye PdM, using data collected by existing sensors, now gives Cameco a continuous stream of information on the condition and predicted future condition of its machines. This will lead to improved sustainability, efficiency and a reduction in downtime.
Using the technique of multivariate analysis, Cameco is poised to make better operational decisions across the business:
- Better data collection means a clearer understanding of what went wrong and stronger insight to avoid repeated issues in the future
- Importantly, it gives a clearer understanding of where and what you should be monitoring to find the best insights on asset health
- Continuous data analysis reveals which forms of condition data are relevant for measuring asset health – and how the relationship between data points can show an upcoming fault
- It provides an opportunity to change operating regimes before they lead to equipment failure. The sooner you know, the sooner you can plan interventions and take advantage of windows in your schedule to apply a maintenance intervention without compromising production
The idea of connected data becomes particularly powerful with the inclusion of operational data. This can help you understand how the way a plant operates affects asset health and overall efficiency.
One example could be how a day’s operations impact vibration. With continuous monitoring, you can discern in real-time whether:
- Vibrations from today’s operations are not causing damage to assets
- Vibration levels are outside of optimal range but excusable as part of one-day operations
- Vibrations from operations are detrimental to asset health and should be stopped
Machine learning systems are used specifically around operations and advanced process control to make optimal operating decisions. They provide more insight into how your operating conditions or regime impact asset health and give businesses new approaches for optimizing their overall operations.
Senseye: Connecting legacy assets to the systems of tomorrow
Senseye research shows that leading metals and mining companies see clear value in connected data. Advanced data infrastructure is almost universal at the largest firms, with 9-in-10 (92%) showing evidence of bringing their data together.
The past five years brought an explosion in the volumes and types of performance sensors that can be retrofitted to legacy machines, allowing the connection of many existing assets to the Internet of Things.
Improving the longevity of legacy assets is crucial for metals mining firms looking to modernize. Deferring the cost of high-value-capital assets offers more flexibility to invest in new developments, while the improved asset reliability keeps operations on track.
To download a copy of Senseye’s Connecting Asset Data at Scale report, please click on the image below.