Mineral processing is a crucial stage in the mining process, involving comminution and concentration stages. Comminution is modeled using various ore variables and operational parameters, representing a complex system. An alternative to simplifying the complexity of these stages is adopting machine learning (ML) techniques; however, ML often requires a substantial amount of data for effective training and validation. The conjoint analysis methodology was used to develop a procedure for discretizing input variables and reducing the data needed for training neural networks, requiring only 77 different scenarios. Using the results from a comminution plant simulator built in Matlab Simulink, neural networks were trained to predict the key output parameters, such as the water consumption, energy consumption, operational parameters, and particle size generated by the plant. The predictive capability of the neural networks was excellent, achieving R² > 0.99 in all cases. The networks were tested with a new set of scenarios to assess their response to values not categorized in the discretization process, achieving R² > 0.98. However, the prediction capability was lost for out-of-range input variables. This approach is attractive for developing easy-to-implement ML tools capable of representing complex systems without needing large amounts of input data, thereby simplifying the modeling process in mineral processing.

In mining, the decrease in the quality of grades has led to further deepening of the existing deposits, determining the existence of a greater quantity of waste, increasing the probability of contact between the waste and the environment, communities and productive soils. For this reason, we propose new logistics to organize mining spaces, using multi–criteria evaluation techniques, risk assessment and the integration of spatial data within the geographic information system (GIS) to determine the optimal locations of industrial mining waste deposits. These optimal locations consider multiple factors such as geological, physical, territorial and environmental characteristics, soil quality and profitability. We also define a sustainable logistic plan for the dynamic distribution of waste using cellular automata, evaluating the interaction of a determined rhythm of waste production, with the dynamic variable of the wind. In conclusion, we have created a new methodology of systematic vision for the planning and territorial ordering of future mining projects.

Induced seismicity resulting from mining activities is one of the major challenges faced by the mining industry. Although such events have been documented for over a century in countries with extensive mining traditions, such as Canada, Australia, and Chile, their impact has intensified over time. This increase is primarily attributed to the greater extraction depths, where elevated stress levels and environmental conditions heighten the likelihood of rockburst occurrences. Seismic events within mines lead to significant human casualties and substantial infrastructure damage, necessitating the implementation of various safety protocols. Among these, seismic indicators are employed to identify periods when high-magnitude seismic events are most likely to occur through the analysis of parameters such as magnitude, energy, time, and decay rate. In this context, the present study aims to utilize the accumulated frictional energy generated by microearthquakes within the Bobrek mine, Poland, as a seismic indicator (variation of frictional energy in time), establishing its correlation with the occurrence of high-magnitude seismic events exceeding the background activity. Thousands of combinations of seismic parameters were tested to maximize the performance of this frictional energy-based indicator, parameters such as moment magnitude, frictional energy, and rock properties. The optimal set of parameters was determined using the Piece Skill Score (PSS) and subsequently applied to the Accumulated Frictional Heat (AFH) methodology. According to the results, the seismic indicator forecasts 86.6% of events with magnitudes M_w ≥ 2.3, with an average forecasting time of 9.76 h, indicating that, on average, these events can be anticipated approximately 10 h before their occurrence.