Metals mining contributes 3 to 4 percent of global CO2 emissions. For open-pit mining, about 45 percent of carbon emissions are typically Scope 1, of which about 35 percent comes from fuel consumed in hauling.
Fuel optimization achieved by harnessing existing data and machine learning can reduce carbon emissions immediately while alternative technologies to diesel for off-highway trucks are developed and scaled.
A proven machine-learning platform enables discovery of correlations and highlights drivers of fuel consumption based on a truck fleet’s past performance by connecting fleet management, enterprise asset management, machine IoT, and other operational data (for example, tire pressure, road layout and quality sensors, and fuel quality). In addition, creation of a digital twin makes it possible to solve for fuel efficiency while maintaining productivity and integrating with both internal and external data sets.
As ore grades decrease and pits become deeper, hauling—and its associated costs—is of greater importance to maintain mine operating expenditures. McKinsey’s experience shows that leveraging proven machine-learning-based solutions, along with a change in management strategy, can improve hauling-fuel efficiency relatively quickly and with limited investment.
Download the full commentary for more insight (PDF–76.3 KB).
