Unplanned downtime is a major issue. Failures in equipment such as car dumpers, stackers, reclaimers, mills and crushers have a disruptive impact and of course a costly consequence in reduced throughput. This advanced AI-based predictive solution enables preventative maintenance, thus reducing plant failures, lost production, spare parts use, labour costs, whilst increasing throughput.

Mining throughput deeply depends on plant machinery functioning at peak uptime, and whilst planned maintenance is an essential process to overall mine performance, unplanned downtime is significantly disruptive and very costly. Equipment such as car dumpers, stackers, reclaimers, mills and crushers all need to be kept running as reliably as possible to maximise tonnage throughput. Many issues can be predicted with the right technology and datasets, combined with advanced AI models, in the fast developing area of predictive maintenance.

This advanced AI platform, which has been used in a large variety use cases across many industries, has been proven to deliver highly reliable predictive maintenance recommendations for large and small mining operations with several successful case studies available. Small customisations are typically used to fit the customer dataset used to power the model, and the solutions uses proprietary artificial neural networks with advanced explainability features and AutoML for scaling. Customers are required to have sufficient sensors and stored data (3-5 years) of the equipment and a downtime accounting system to record and store malfunctions. The team will then work with the customer on data integration and input into the system.