Razor Labs’ DataMind AI transforms mining operations by predicting failures before they happen.
Prevention, as they say, is better than cure. This is especially true on a mine site, where predictive maintenance is among the most effective approaches to ensuring operations run smoothly.
While preventive maintenance offers benefits such as reduced risk of equipment breakdowns and decreased downtime, it can be more labour-intensive, resulting in increased staffing and operating costs.
This is why more and more mining companies are turning to predictive maintenance solutions.
Predictive maintenance analyses real-time sensor data to detect failure risks early, enabling proactive repairs before breakdowns occur.
Razor Labs is a pioneer in AI-driven predictive maintenance. Its fully automated system has optimised asset reliability for leading mining companies since 2016.
Described as a ‘single out-of-the-box solution’, Razor Labs’ DataMind AI predictive maintenance system is designed to eliminate unplanned downtime, increase productivity, and enhance safety.
DataMind AI integrates sensor data from mining equipment, using neural networks to detect hidden failure patterns. This enables operators to predict and prevent costly breakdowns with precision.
“We first partnered with a Tier 1 miner in Australia to develop AI-driven failure prediction models using time-series data and other insights from its existing systems,” Razor Labs chief business officer Tomer Srulevich told Australian Mining.
“We start by analysing key maintenance KPIs (key performance indicators), particularly the planned versus unplanned maintenance ratio, to pinpoint where reliability can be improved.
Image: Razor Labs
“We assess whether the main constraint is in the mine or the plant, identifying critical bottlenecks that restrict throughput. By mapping single points of failure, we help sites reduce downtime, streamline maintenance, and eliminate production constraints for maximum efficiency.
“If the client has existing sensors, we can easily integrate them into DataMind AI. If not, we map optimal placements for full asset coverage, ensuring seamless data collection.”
Built to withstand tough environmental conditions, DataMind AI’s platform integrates sensors for temperature, pressure, current, vibration, and oil, providing comprehensive real-time equipment monitoring.
The platform also includes a camera for vision-based analysis, enabling material monitoring and belt drift detection, ensuring 24–7 visibility with no sensor gaps.
“DataMind AI’s sensor fusion technology pinpoints failure root causes and delivers AI-driven prescriptive actions to prevent breakdowns,” Srulevich said. “Integrating data from multiple sources and various sensor types is one of the biggest challenges in mining. Fragmented systems, handheld devices, and isolated sensors often miss critical failure indicators, creating blind spots in equipment monitoring.
“Unlike traditional monitoring systems, DataMind AI’s sensor fusion combines vibration, temperature, pressure and oil analysis – offering a multi-layered diagnostic advantage.”
DataMind AI is deployed by a pre-approved and inducted site integrator, eliminating the need for new vendors. With minimal training, the integrator can execute the installation efficiently, ensuring rapid deployment without complex model calibration.
“Once it’s up and running, the system immediately begins generating predictive insights,” Srulevich said.
“Some of the equipment installations can even be done while the machine is operating. We continuously sharpen our process to make sure the implementation is as short and efficient as possible.
“DataMind AI has already helped mining operations across Australia, identifying critical failures in both iron ore and coal sites.”
At the iron ore mine, DataMind AI detected rapid deterioration of a critical pump bearing overlooked by manual inspections.
Envelope demodulation acceleration increased six to nine times compared to baseline, with AI sensor fusion compensating for operational deviations. The system flagged the issue as ‘critical’, enabling timely bearing replacement during scheduled downtime and preventing costly secondary damage.
At the coal mine, DataMind AI identified fluting in a conveyor motor bearing caused by electrical currents passing through bearings, reducing their lifespan and increasing maintenance costs. Early detection allowed proactive maintenance, avoiding unplanned stoppages and extending equipment life.
“These outcomes underscore DataMind AI’s real-world impact across diverse mining environments and commodities in Australia,” Srulevich said.
“Our ability to aggregate vast data sets and transform them into actionable insights is unmatched in the industry.
“We continue to push the boundaries of AI-driven predictive maintenance, setting new standards in mining efficiency.”
This feature appeared in the April 2025 issue of Australian Mining.