Customer Case Study:

Digital Twins in Mining Operations and Maintenance

Global Mining and Resources Customer

This multi-national customer leads the world in producing critical natural resources. They use the best technologies, responsible mining methods, and their experienced team to stay on top.

They’ve been in the business for many years, have many high-quality reserves, and have expanded over multiple years. This allows them to meet the global demand for this resource effectively.

They run many low-cost mines and processing plants. Their vast and reliable distribution network, including marine terminals, numerous distribution points, and many owned or leased railcars, allows them to supply their products to about 40 countries.

Their operations are backed by a digital transformation program to increase production and cut costs using automated mining and digital technologies. This program prioritizes safety, efficiency, cost-effectiveness, flexibility, and environmental and social responsibility.

In 2022, they made more use of automated mining, boosting production by about 50% compared to the previous year. They also grew their predictive maintenance platform, using advanced technology such as XMPro to foresee and prevent asset failures, with mobile sensors improving their monitoring abilities.

The Challenge:

In 2019, the client undertook an initiative focusing on predictive maintenance and operations. The goal was to enhance production and operate at optimal capacity across their mining ventures. Multiple digital twin scenarios were considered, and the reduction of underground long conveyor downtime was chosen as the pilot project for XMPro iDTS, guided by XMPro’s Use Case Prioritization Matrix framework. The objective was to cut conveyor downtime for a specific failure mode by 30% over a 120-day trial period.

The Approach:

The solution used XMPro iDTS to amalgamate real-time data from various sources including real-time sensors, historians, and maintenance management systems. Expert guidelines from seasoned maintenance teams, some with over 40 years of experience in soft rock mining, were used as the benchmark for recommendations that assessed the real-time data every two seconds. This experiment was carried out at a featured mine, where the system oversaw more than 50 conveyor belts spanning over 80km underground during the preliminary evaluation. The initial version was up and running, starting to deliver value within 30 days.

“Tube Map” view of underground conveyors with predictive alerts markers

At the start of the pilot

120 days later

The Results:

The initial pilot was very successful, reducing downtime by over 80%. In the four-month proof of concept period, they prevented 60 hours of borer downtime as a result of conveyor stoppage, equivalent to approximately 14K tons of product. They identified a further 124 hours of preventable downtime, amounting to roughly 30K tons of product.

The XMPro Recommendations effectively captured expert knowledge on maintenance best practices, transitioning from ad-hoc business intelligence-style Excel spreadsheet analysis to continuous scrutiny and alerts. This provided reliability engineers with constant insights to prescribe the necessary actions. The recommended actions were then incorporated into their predictive and prescriptive maintenance procedures.

Following the success of the pilot, XMPro was implemented across the other mining sites to tackle a variety of operational issues across eight asset classes. With 32 use cases, and sending 42 million messages per day, the business reaped over 10X return on investment on the project.