Latest news with #predictiveMaintenance
Yahoo
26-05-2025
- Business
- Yahoo
Mining Smarter: Inside Razor Labs' Award-Winning Approach to Predictive Diagnostics and Safety
In this exclusive interview, Tomer Srulevich, Chief Business Officer at Razor Labs, shares how the company's award-winning DataMind AI™ platform is transforming mining operations. Srulevich discusses the journey from overcoming data integration challenges to delivering predictive, AI-driven maintenance that reduces downtime, boosts safety, and maximizes asset life. Mining Technology: Congratulations on winning the Innovation award in the Equipment Diagnostics category! What does this recognition mean for Razor Labs and your team? Tomer Srulevich: It's a proud moment for the entire team. This award recognizes the years we've spent developing, deploying, and refining AI solutions tailored for mining's unique challenges. It's not just a technology win — it's a validation of the impact we're delivering to the field. Our work eliminates blind spots, prevents breakdowns, and makes sites safer and more efficient. This acknowledgment also energizes us to keep pushing the boundaries of what AI can do in heavy industry. Mining Technology: Can you elaborate on the vision behind DataMind AI™ and how it aligns with Razor Labs' overall mission in the mining industry? Tomer Srulevich: From the start, our vision was to replace reactive maintenance with predictive, AI-driven action — not just alerts, but full root cause insights and prescriptive steps. DataMind AI™ is built to be a single out-of-the-box solution that integrates with existing infrastructure or stands on its own. We aim to reduce unplanned downtime, maximize asset life, and give teams total visibility across critical equipment — whether that's pumps, mills, crushers, or conveyors. It's our way of embedding smart, real-time decision-making directly into site operations. Mining Technology: DataMind AI™ clearly demonstrated significant cost savings and operational efficiency. How do you measure the success of your AI solutions in real-world applications? Tomer Srulevich: We measure success where it matters most: downtime avoided, dollars saved, and reliability increased. For example, at an iron ore site, we identified rapid pump bearing deterioration that traditional inspection missed — saving the customer secondary damage and costly unplanned shutdowns. In another case, we uncovered electrical fluting in a conveyor motor bearing at a coal mine, enabling early intervention that extended the equipment's lifespan. These outcomes saved hundreds of thousands of dollars — and just as critically, they prevented safety risks and disruptions to throughput. Mining Technology: What were some of the key challenges you faced while developing DataMind AI™, and how did you overcome them? Tomer Srulevich: One of the biggest hurdles was integrating fragmented, siloed data from vibration sensors, SCADA, and handheld tools into a coherent, AI-readable format. We tackled this by building a Sensor Fusion architecture — combining temperature, pressure, current, vibration, oil, and visual data into a unified model. Another challenge was adapting to low-speed equipment like kilns and stackers, where traditional vibration tools struggle. Our custom envelope demodulation and edge computing made it possible to detect issues at sub-100 RPM — something the market lacked until now. Mining Technology: How do you see real-time monitoring evolving in the mining industry, and what role will Razor Labs play in that evolution? Tomer Srulevich: Real-time monitoring is evolving from basic alerting to automated diagnostics with prescriptive actions. It's not enough to know something is wrong — you need to know what to do about it. DataMind AI™ is already leading that shift by identifying not only failure modes but root causes, and suggesting targeted corrective actions. We're pushing toward an environment where maintenance is intelligent by default, and where downtime is scheduled — not imposed. Mining Technology: Can you discuss how DataMind AI™ differentiates itself from traditional monitoring methods, particularly in terms of technology and data analysis? Tomer Srulevich: Traditional methods rely heavily on threshold breaches or isolated sensor readings. That approach is reactive and limited. DataMind AI™ uses deep neural networks, multi-sensor fusion, and envelope demodulation to detect early, hidden signs of failure. For instance, at one site, our system detected fluting in a bearing due to stray electrical currents — a subtle issue that vibration velocity alone wouldn't catch. Unlike legacy tools, we analyze vibration in the acceleration domain and fuse it with electrical current and pressure data for far greater accuracy. Mining Technology: How do you ensure that your team stays updated with the latest advancements in AI and machine learning to continuously improve DataMind AI™? Tomer Srulevich: We maintain a continuous feedback loop between our R&D team and field deployments. Every detection event, every resolved failure, and every false positive informs how we refine our models. We also collaborate with academic institutions and maintain internal research programs on semi-supervised learning, anomaly detection in rare-event environments, and edge AI deployment. And critically, we stay grounded — we spend time on-site, shoulder-to-shoulder with the teams using our product. Mining Technology: What feedback have you received from clients regarding the implementation of DataMind AI™, and how has it influenced your product development? Tomer Srulevich: Clients consistently appreciate our precision and clarity — not just saying 'there's an issue,' but showing what the issue is, where, and why. One client told us, 'It's like having an expert sitting in the control room 24/7.' That feedback has led us to enhance our visual diagnostics — things like showing the actual vibration signature or video frame triggering the alert. We've also improved integration with SAP, allowing sites to generate maintenance notifications directly from our dashboard. Mining Technology: Looking ahead, what are some strategic goals for Razor Labs in the next 3–5 years, particularly in innovation and technology? Tomer Srulevich: Strategically, we're expanding DataMind AI™ across mobile assets like haul trucks and shovels, while deepening our edge analytics capabilities. We're also advancing in process optimization — using AI not just to detect faults, but to suggest settings that improve efficiency or reduce energy use. Another goal is to make AI-driven reliability management more autonomous and intuitive — reducing dependence on site experts and enabling consistent best-in-class maintenance across all sites, regardless of their internal resources. Mining Technology: Safety is a critical concern in mining operations. How does DataMind AI™ contribute to enhancing safety for workers on-site? Tomer Srulevich: Safety is embedded in our mission. By catching faults early, we eliminate risky emergency interventions — like climbing on moving conveyors or opening live panels. For instance, in a centrifuge drum imbalance case, our system flagged an issue before manual testing could catch it. The site avoided both mechanical failure and the dangerous physical inspection that would've been required during operation. We're turning reactive firefighting into calm, preemptive action. Mining Technology: Can you share any upcoming projects or innovations that Razor Labs is currently working on that you believe will further impact the mining industry? Tomer Srulevich: We're currently developing 'virtual sensors' — algorithms that estimate unmeasurable parameters like wear state or residual life based on fused input. And on the platform side, we're enhancing self-service model tuning, allowing users to adapt diagnostics to specific local conditions without retraining the AI. Mining Technology: What partnerships or collaborations do you see as essential for advancing the capabilities of DataMind AI™ and enhancing its impact on the mining industry? Tomer Srulevich: Strong partnerships with OEMs and EPCMs help us integrate deeper into the equipment ecosystem. But just as important are our on-the-ground relationships with site maintenance teams — they're the ones validating alerts and applying corrective actions. We also collaborate with software vendors like SAP and BI platforms to ensure seamless integration into our clients' operational workflows. The future of mining reliability lies in open, interoperable ecosystems — and we're committed to building them. Mining Technology: Thank you, Tomer, for sharing your valuable insights into Razor Labs' journey and the transformative impact of DataMind AI™ on the mining industry. Congratulations again on your well-deserved recognition, and thank you for your commitment to driving meaningful change in the industry. E-mail: pr@ Links Website: "Mining Smarter: Inside Razor Labs' Award-Winning Approach to Predictive Diagnostics and Safety" was originally created and published by Mining Technology, a GlobalData owned brand. The information on this site has been included in good faith for general informational purposes only. It is not intended to amount to advice on which you should rely, and we give no representation, warranty or guarantee, whether express or implied as to its accuracy or completeness. You must obtain professional or specialist advice before taking, or refraining from, any action on the basis of the content on our site. Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data
Forbes
09-05-2025
- Business
- Forbes
AI Unearths New Potential In The Mining Industry
AI in mining Mining might not be something you think about daily, but every physical product we use and depend on is dependent on natural resources that come from and are extracting from the Earth. AI is increasingly being utilized in the mining industry to improve efficiency, safety, and sustainability. So let's dig into it. Mining operations are performed in unforgiving and hazardous environments. AI systems are helping to improve overall efficiency and safety by removing the human from the most dangerous environments and augmenting human capabilities. There is increasing use of AI to power autonomous trucks, drills, and loaders used in mining operations. These AI-driven vehicles can operate in hazardous environments with precision, improving safety and productivity. Companies have developed autonomous mining vehicles that are already being used in large-scale mining operations. When equipment operates in challenging environments, they need continuous maintenance. However, maintenance takes critical equipment offline and consumes resources. Being more precise with maintenance means increased uptime for expensive and necessary equipment and significant cost savings. AI-driven predictive maintenance systems monitor the health of mining equipment by analyzing sensor data to predict when machinery might fail, improving equipment reliability and lifespan. There are many processes involved in extracting resources from the ground and making them usable and accessible for their applications. The more that these processes can be made more efficient, safer, less environmentally impactful, and more reliable, the more that those benefits will be passed upstream to those who consume those resources. AI is used to optimize various mining processes, such as crushing, grinding, and flotation, by analyzing real-time data and adjusting parameters to maximize efficiency. AI-driven process control systems improve throughput, reduce energy consumption, and enhance overall operational efficiency. AI systems are also helping to plan mining operations and estimate resources as part of mining processes. AI improves resource estimation by analyzing geological data to provide more accurate assessments of available resources. This enables better mine planning and more efficient extraction of minerals. AI-driven mine planning tools help optimize the layout of mines, reducing waste and maximizing resource recovery. AI-driven mine planning tools help optimize the layout of mines, reducing waste and maximizing the overall resource recovery. AI systems are also being used as part of ore grade prediction and exploration. These systems analyze geological and sensor data to predict the location and quality of ore deposits, using patterns in seismic data, drill hole information, and satellite imagery to overall improve the accuracy of mineral exploration and reduce the time and cost involved in finding new resources. As mining environments are not the most friendly for human activity, AI systems can be put to good use keeping mines safe and well operated. AI is increasingly used to help with overall safety monitoring and incident prevention by analyzing data from sensors, cameras, and wearable devices, helping predict and detect potential hazards, such as rock falls or gas leaks, or equipment failures. The AI systems can alert workers and supervisors so that they can take preventative actions, reducing the risk of incidents and energy and injuries Mining operations are also very energy intense. AI systems help optimize energy consumption in mining operations by analyzing usage patterns and identifying opportunities to reduce energy waste. These energy management systems can optimize power usage across different processes, lowering operational costs and reducing the environmental impact of mining activities. The range of environmental impacts of mining operations include air and water quality, water usage, waste management, managing mine outputs, impacts on underground and above-ground land and environment, and human and animal impacts. Mining operations must comply with a range of regulations and compliance activities to both keep their operations safe and minimize those environmental impacts. AI can monitor environmental factors such as air quality, water usage, and waste management to ensure compliance with environmental regulations. AI-driven systems can detect deviations from permitted levels and recommend corrective actions, helping mining companies minimize their environmental footprint and avoid fines. Mines generate not only useful outputs but also other side-effects of extraction such as waste water and materials that need to be managed such that they don't in themselves cause problems. Called tailings or tails, these are the materials left over after the process of separating the valuable parts of the extraction from the other parts. These tailings are often stored and managed in order to be reliably and responsibly disposed of. AI enhances tailings management by monitoring the stability of tailings dams and predicting potential failures. AI-driven systems analyze data from sensors embedded in the dams, such as pressure and moisture levels, to detect early warning signs of dam breaches. This helps prevent catastrophic failures and environmental disasters. While mining operations are just started with applying AI to all these areas, we can foresee a future where not only we get access to the much-needed resources that power and support our daily lives, but also continue to do so in a safe, efficient, and effective way.
