Cases and Applications
In the survey conducted by [10] on some mining companies in Australia, the conclusion was that PdM, Big Data, IoT, and Data Analytics create safe work environments, enable better maintenance practices, and reduce maintenance costs. More than one-third of the total mining costs are destined for maintenance processes, which makes it the highest controllable cost. The majority of failures do not occur instantaneously, and normally there are few signs of degradation processes or trends that indicate the transition from normal status to failure [11]. Supported by the emerging technologies, miners can collect and assess valuable data and make more reasonable maintenance decisions.
Digital Twin for Intelligent Maintenance
The smartest level of advanced analytics for maintenance can be achieved through the implementation of Digital Twins (DT). In this advanced approach, all information collected about an individual physical asset or combination of assets in a system is used to maximize asset utilization and optimize its performance. The historical data used to model a DT includes operational conditions under which it has been used, its configuration, maintenance events, and other exogenous conditions, among others. The DT is designed to enhance human effectiveness, productivity, and capacity by focusing on more strategic business issues [1].
In mining applications, the models start by guiding "design limits" and, then, are continuously updated and learn to efficiently mirror the asset under different operational scenarios and related variations - ore types, temperature, weather conditions, air quality, moisture, load, weather forecast models, and more. Combined with advanced technologies, DT models can reach the next level of optimization, control, and prediction, delivering more accurate results for asset performance, reliability, and maintenance. Additionally, by using the sensor data, models enable mitigation of threats and unplanned downtime with the evaluation of different scenarios, understand trade-offs, improve efficiency, and ultimately perform PdM. DT solutions are the future of mining asset management, and the number of successful implementation cases are constantly growing. Figure 7.8 depicts the use cases of DT models across the entire asset's lifetime, starting with the design phase.
According to Kevin Shikoluk, a global strategic marketing leader for Digital Mine at GE, DT goes beyond the prediction that an equipment or component will breakdown. DT models show what will happen if the problem is not fixed. They allow its users to test different changes in real time before executing them in the plant [12,13]. Some mining giants like Rio Tinto, Anglo American, ВНР, and Newmont are embracing DT solutions not only for individual assets but to digitalize the entire operation [14-16].
