Abstract:
The mining industry still finds it difficult to turn large streams of operational data into decisions
that are both fast and useful. At times the reports or dashboards arrive too late. By the time the
numbers are available, the people on site may already need to act. This delay can bring
production losses, extra safety concerns, and higher costs (Rojas et al., 2025). Artificial
intelligence seems to help in some areas, such as predictive maintenance and equipment
reliability (Sahoo et al., 2019; Abdi et al., 2008). Even so, its role in mining is not very broad.
In many cases the data is noisy, systems fail to connect smoothly, and the models can be
difficult for people to trust or explain (Landre Júnior et al., 2022). In this study we put forward
MineMind, an AI-based model for performance management in mining. The idea is not only
to monitor or report but also to support decisions that should be more proactive and adaptive.
To develop the model, the work follows two paths. One path looks at benchmarking. It reviews
how AI has been used in other industries, with energy being the main point of reference, and
pulls out lessons that could make sense for mining. The other path turns to product
development. Here the focus is on building MineMind with elements like real-time monitoring,
natural language interaction, and AI-driven recommendations. For guidance, the research
makes use of the Digital Transformation Pyramid (Westerman et al., 2014). Yet the framework
is not taken as is. It is adapted to fit the mining setting. The results suggest that combining
benchmarking with product development may help MineMind support mining companies in
strengthening control, improving safety, and running operations more efficiently. On a wider
note, this study also adds to technology management by showing one possible way to apply AI
to mining performance management. It does not aim to close the discussion but rather to
provide direction.
