and local practices. “We then spend roughly three to four weeks integrating the system, return to site for implementation and training, and focus on optimising that initial area. Once it’s working, we progressively expand into the rest of the operation until the entire mine is covered—including the processing plant, where capacity and blending decisions directly inform daily planning. Initial value typically appears within about two months, after which we expand horizontally across the operation.” The company took roughly four years to build the Hivekit platform, which is in production across several sites. In August Hivekit launched OPS.AI, a new technology that proactively coordinates and optimises mine operations, identifies bottlenecks, answers operational questions and executes authorised directives. “The goal is to help supervisors plan each shift more effectively by giving them a much richer, real-time view of what’s happening across the site,” explains Hempel. Early tests against historic mine operations data showed a 21% improvement in compliance to plan, alongside significant improvements in the use of existing resources. “We saw 7% and 11% improvements in ounces delivered to the mill across two sites, which is our core metric. We also saw a 15% improvement in fleet use and 21% improvement in compliance to plan. The key point is that these gains come from better use of the equipment and people mines already have—the operations did not need to buy more equipment or hire more people. Using the technology, the mines are simply able to get more out of the existing assets, which is increasingly valuable when every ounce matters,” says Hempel. OPS.AI achieves these efficiency gains by connecting the mine’s strategic and production plans with day-to-day operational execution. This holistic, real-time view of the operation allows OPS.AI to plan around actual resource availability, identify correlations, constraints and bottlenecks, and adjusts shift goals and production cycles to real-world conditions. The technology can assign tasks, cycles and targets, and allocate people, vehicles and other resources accordingly. “The real value comes from the operational scheduling layer. With OPS.AI, supervisors can generate an optimised shift plan, re-plan throughout the day as conditions change, and continuously optimise targets, people, resources, and inventory. Beyond this, supervisors can interact directly with OPS.AI — for example, asking why a truck is down and getting an answer based on the underlying operational data. Dashboards and analytics then provide the broader context,” he explains. Partnerships in Canada and Peru For the past three years the technology specialist has been working closely with two mining houses – one operating an underground gold mine in Canada, and with a surface copper and silver mine in Peru. “We started building OPS.AI about two years ago, which was really the earliest point at which the underlying technology became viable. We developed the platform in collaboration
Hivekit is an operations platform for modern mines, which combines real-time visibility, AI-assisted planning, and short interval control into one connected system.
Hivekit architecture diagram.
November 2026 | www.modernminingmagazine.co.za MODERN MINING 19
Made with FlippingBook flipbook maker