Search Documents

By Abdallah Khair, Nathalie Risso, Arie Herrera, Victor Tenorio

This study introduces a machine learning strategy to optimize icy regolith extraction from the Shackleton-de Gerlache Ridge using autonomous mining equipment. The model incorporates key parameters—distance, road gradient, vehicle load, speed, recharge logistics, and regolith water content—into a cost framework for excavation, hauling, and dumping. Regression models estimate energy and time requirements; classification predicts terrain navigability; clustering groups craters for efficiency; and K-Nearest Neighbors (KNN) determines optimal routing and scheduling. The system adapts to a six-month daylight cycle to maximize productivity and minimize costs, enabling intelligent planning for sustained lunar resource extraction under extreme environmental and logistical constraints.

Feb 22, 2026

By Francisco J. Sotillo

Using new ecological-natural-organic depressant for gangue and pyrite with antibacterial properties (control of acid mine drainage) with new improved xanthate collectors using rheology additive not chemical reacting with xanthate resulted in better sulfide flotation. Rougher-scavenger lead flotation tests were conducted increasing the concentrate grade from 7.07% Pb to 13.44% Pb, enhancing selectivity. Lead recovery increased from 64.08% to 79.38% and 80.40% using only the new depressant, and improved xanthate Z-11 and Z-4, respectively. The results demonstrated that the new depressant was selective and efficient, and the new improved xanthates stronger more selective collectors.

Feb 22, 2026

By Giorgio De Tomi, Rondinelli Sousa

Whilst all public mineral reporting codes require classification of mineral resource and reserves according to the level of geological knowledge and confidence of the estimates, companies are not required to disclose how mineral estimates and associated cashflow projections respond probabilistically to the incorporation of different types of risk events. As the level of expectations and scrutiny of investors and stakeholders continue to increase over the years, it will be essential for companies to adopt an integrated framework for sensitivity analysis to increase corporate value through better capital allocation and reduce potential losses associated with poor investment decisions. Failure in fully capturing the impact of unexpected incidents in mineral estimates and financial modeling could be extremely harmful to the business, affecting the company’s reputation, and negatively impacting the relationship with stakeholders.

Feb 22, 2026

By Bright Oppong Afum, Kojo Boakye, Kwaku Boakye

This research developed and simulated a novel multi-indicator Fatigue Detection and Alert System (FDAS) designed to enhance fatigue detection accuracy in mining environments. The system integrates a comprehensive set of inputs: eye blink rate, eye closure duration, yawning frequency, distraction levels, Pre-shift Check-in Scores (Self-report) and Electroencephalography (EEG) Levels. Using MATLAB’s Fuzzy Logic Rule Viewer (FLRV) and Simulink, simulations demonstrated the FDAS’s robust ability to process these diverse physiological and behavioural indicators accurately. It effectively classified operator fatigue into low, moderate, and high fatigue levels, consistently providing appropriate outputs and corresponding intervention recommendations across various mixed-signal scenarios. The system achieved a calculated accuracy of 85.19% across 27 test entries, with a standard deviation of approximately 1.8459, confirming its reliability for real-time fatigue monitoring. These findings highlight the significant potential of this fuzzy logicbased approach to proactively identify early signs of fatigue, substantially enhancing safety and operational efficiency in demanding mining operations.

Feb 22, 2026

By Jim McKinley

With a growing number of mines reaching closure and seeing closure costs materialize, regulators, lenders, and banks are observing projects that fail economically by miscalculating these costs. Mine closure costs are often categorized as a type of liability which is defined as the money owed by the company to offset effects from the project, and the money available to pay for those debts (e.g., bond). Often, decisions made before Day 1 of mining (e.g., co-disposal, dry stack tailings, waste segregation) can significantly change the bond calculation. Once the bond is calculated and funded, operators are encouraged to conduct closure research and use applied science for progressive bond release prior to final closure. However, this author has observed few instances of financially significant progressive bond release, as there are typically no robust procedures for bond return, either before or during closure activities. As a result, many companies decide that entering a state of Care & Maintenance (C&M) is less risky than attempting to fund closure activities with an unknown bond return mechanism. Although perhaps less risky in the short-term, the primary long-term risk with entering C&M is that it increases the chances of the bond not covering the cost of closure and the chances of the government/regulatory agency inheriting the mine closure. Bonds are typically reevaluated on a regular cycle (e.g., five years) so every mining company goes through this financing challenge eventually.

Feb 22, 2026

By Ali Soofastaei

Mining operations are increasingly data-rich yet decisionconstrained. Fragmented systems, variability in geology and operations, and sustainability obligations create multiobjective trade-offs that classic analytics cannot reconcile at operational tempo. This paper proposes an end-to-end framework that fuses digital transformation practices (governed data platform, semantic layer, feature store, and MLOps) with generative AI (retrieval-augmented generation, tool-use, and agentic workflows) and decision intelligence (simulation/digital twins and multi-objective optimization). We formalize sustainability-aware objectives, define a practical stack and governance controls, and demonstrate how an AI “decision copilot” can surface options, quantify trade-offs, and document rationale. Two figures illustrate a production-grade architecture and a Pareto frontier for cost–emissions trade-offs; two tables operationalize decision-domain mapping and a transformation maturity rubric. The approach is product- and company-agnostic.

Feb 22, 2026

By M. E. Rossi

In mineral resources, geologic and mining data are acquired, processed, analysed, interpreted and modelled, eventually serving as basis to generate predictions and production plans. This is a complex process, and the resulting predictions are impacted by uncertainties related to the data itself and the processing methods and techniques used. They are also impacted by the subjective decisions made by the modeler(s) at each step of the way. Knowledge, biases, skills, and prior experience come into play in the long mineral resource evaluation development workflow. It is meaningless to discuss uncertainties without evaluating the associated risks that those uncertainties imply. It is important to quantify and assess uncertainty, and to evaluate and mitigate its consequences. This paper discusses (1) conceptual and practical sources of uncertainty; (2) the utopia of an “objective” uncertainty; and (3) the quantification and public reporting of uncertainty using geostatistical conditional simulations. The objective of this paper is to postulate that the industry is running out of reasons and excuses to ignore uncertainty and risk in project evaluation. This is done in two parts, with this, the first one, presenting a more conceptual discussion. Part 2 will present case studies from different types of mining projects and operations showcasing the impact that risk assessments can have in a mining project.

Feb 22, 2026

By Lina Ospino Ballesta, Oscar J. Restrepo Baena, Juan Diaz Martínez

Artisanal and small-scale underground gold mining in Colombia plays a significant role in the country’s economy, but it faces persistent occupational hazards that threaten worker safety. Despite mining economic contributions, outdated technologies, limited risk management practices, and insufficient regulatory enforcement have led to recurring workplace accidents, many of which have been fatal. This study presents a systematic review of mining-related accidents over the last decade. Through geo-referenced analysis, high-risk municipalities were identified, and the frequency of accidents was correlated with the mining legal status and operational practices of these municipalities. The findings underscore the urgent need for evidencebased interventions that are tailored to the country’s unique mining context. Ultimately, this research contributes to the global discourse on mining safety by highlighting practical pathways to reduce occupational risks and promote safer, healthier workplaces in resource-constrained environments.

Feb 22, 2026

By Corey Vaughan

Unplanned shutdowns from abrasive wear part failure in crushers, chutes, and slurry pipe assets can cost millions in lost production and reactive maintenance. 2Censor patented sensor technology enables real time abrasive wear monitoring of mineral processing equipment, turning passive assets into smart, data driven components. Wireless sensors embedded in liners and pipework transmit wear data live to operators, allowing for predictive maintenance, safer scheduling, and optimized asset life. This paper will present global case studies showing reductions in downtime, increased throughput, and improved maintenance accuracy. The system is ready to retrofit, neutral to original equipment manufacturers, and proven across more than twelve countries in abrasive wear liners and slurry pipe applications.

Feb 22, 2026

By Derek du Preez, Andrew Hall, Timothy M. Pelech

Net Present Value (NPV) derived from Discounted Cash Flow (DCF) analysis remains the dominant metric for evaluating mining projects despite its well-documented inability to capture managerial flexibility under uncertainty. DCF assumes that all risks are perfectly priced by financial markets. This is an unlikely assumption for volatile commodity prices, geological variability, or the array of mining technical and operational risks present on all mine sites. Real Options Analysis (ROA) was developed to remedy this limitation by explicitly valuing the ability of managers to defer, expand, scale, switch, or abandon investments as new information emerges, and to capture upside and mitigate downside risks. While ROA is philosophically correct and routinely shows an uplift in project value compared with DCF, its practical application at industrial scale is severely constrained, ROA faces the reality of a combinatorial explosion in complexity of decision trees, Monte Carlo simulations and other models required when multiple interdependent or sequential options exist across numerous mining zones and time periods. The AMC Hill of Value® (HoV) process adopts the same intellectual foundation as ROA, that flexibility has economic value, but operationalizes it through a far more robust and scalable architecture. It is built on parametric relationships, rapid scheduling, and multi-objective evolutionary optimisation. HoV can evaluate millions of compound, interdependent real options in a single optimisation run, delivering resilient, adaptive strategies that preserve managerial flexibility while simultaneously optimising NPV, mine life, risk exposure, and other corporate objectives. This paper demonstrates that HoV is the practical, industrial-scale embodiment of real options thinking in mining: it discovers, ranks, and translates the most valuable real options into strategic mine plans, overcoming the scalability barriers that have prevented widespread adoption of classical ROA.

Feb 22, 2026

By A. J. Richins, M. G. Nelson, R. Dunne

From the inception of flotation in the late 1800s, coarse particle recovery has been a challenge to practitioners. By the early 1900s, flotation processes included oil flotation, skin flotation, agglomeration flotation, and froth flotation, and for each there was a proliferation of flotation equipment. Ensuring good coarse particle recovery included desliming of the feed and proper conditioning were required. Froth flotation was at first used mainly to treat finer materials from grinding circuits. The maximum size for coarse particles was from 3.18 to 4.78 mm. In froth flotation, large particles with high specific gravity tend not to be recovered. Coal and minerals of low specific gravity and flaky minerals such as graphite, mica, and talc can be floated in larger sizes. In the early 1960s, the phosphate industry began using column cells to recover coarse particle by froth flotation, but early columns suffered from poor bubble generation and sanding at the flat base of the column. Recently, improved coarse particle recovery has been the object of new cell designs—the Eriez HydroFloat™, the Jord NovaCell™, the FLS coarseAir™ cell, and the Cidra P29 integrated system. The paper will follow the development of coarse particle flotation with a focus on the development, laboratory assessment, and operation of the new flotation equipment and technology designed to recover coarse particles.

Feb 22, 2026

By Nelson Alonso Portal Ravichaua, Melissa Esther Mendoza Tirado, Raúl Arévalo Saavedra

This study analyzes the implementation of an integrated strategy for environmental education and community awareness in Cajamarca, Peru—an Andean region affected by climate variability and the public perception of miningrelated water shortages. In collaboration with the Interactive Water and Earth Museum and the Newmont ALAC Foundation, three-month campaigns were conducted in schools and local communities. Key topics included the water cycle, climate change, and water availability, linked to urban infrastructure projects that improved flow rates through strategically located tubular wells tapping underground aquifers. The initiative addressed the challenge of communication in a context where mining is often blamed for scarcity, fostering scientific understanding of contributing factors. The study highlights how science-based outreach and institutional collaboration can strengthen resilience and promote sustainable water stewardship in mining-influenced regions.

Feb 22, 2026

By Scott O’Brien, Richard Williams, Camilo I. Mejias, Elias Godoy, Javier Sierra

High-rate thickeners play a critical role in tailings management by enhancing water recovery and increasing solids concentration. However, these units face operational instability due to fluctuations in feed solids, flow rate, and flocculant response typically managed in manual mode with delayed corrective actions. This work presents an adaptive automatic control algorithm designed for dynamic thickening environments. The algorithm integrates real-time measurements of feed density, rake torque, bed level, and turbidity (overflow), optimizing the underflow slurry density while minimizing suspended solids in the overflow. The proposed control strategy is modular, simple, scalable, and applicable to both existing and new thickening systems, offering a robust alternative to manual operation in increasingly demanding metallurgical environments.

Feb 22, 2026

By Klaydison Silva, Pedro Silva, Letícia Damet

The current situation of the global mining economy is increasingly focused on the utilization of materials previously discarded in the mining process. The waste from the past, through new processing routes, is gaining new contours for use as high-value-added products. The strategic mine planning process has consequently had to adapt to the new economic reality. In standard scenarios, the focus was primarily on mine sequencing aimed at the highest economic return, with all infrastructure allocated as a necessity of mining. This work aims, in strategic planning, to configure and select the best composition of mining structures. The entire Mine-Plant-Waste-Tailings-Products chain will be considered as a factor in the sequencing of a multimine complex, to achieve the most sustainable utilization of the current mines’ waste. The sequencing is based on concurrent flows of process configurations and capacities, structure sharing, and flexibility of feed types. These factors will guide the sequencing, which will be constrained by components such as waste and tailings disposal rates, lithology feed relationships, and especially product composition through market specifications, all while respecting product shipping capacities.

Feb 22, 2026

By Patricio Vergara Lara, Jorman Barrera

The selection of the production level elevation is one of the most critical strategic decisions faced by large-scale block or panel caving projects. This elevation largely affects project timelines, capital expenditures, resource recovery, and may even constrain the mine’s maximum production capacity. Over the years, many computational tools have been developed to assist in this decision, most of them based on the same principle of maximizing the project value given its production level elevation. However, for large and inclined deposits, or those extending significantly in the vertical dimension and where extraction cannot be achieved from a single level, traditional selection techniques involving multiple production horizons often lead to suboptimal results. This paper proposes a methodology ensuring that the selection of production level elevations in a multiple horizons (“lifts”) case is globally optimized. Input parameters include economic and cost variables, maximum extractable column height, and deepening costs. The methodology is demonstrated through a real case study to illustrate its potential impact on project value.

Feb 22, 2026

By Oscar Cabello, Luis Mendieta, Maribel Guzman

Mineral flotation processes generate large volumes of tailings, which must be disposed of efficiently and safely. In addition to dry stack disposal, the possibility of reusing these tailings as aggregates in shotcrete is an interesting alternative. Tailings samples from a mining operation in southern Peru were characterized by granulometric analysis, X-ray diffraction and Scanning Electron Microscopy. Concrete samples were obtained with different tailings formulations. The samples were subsequently subjected to geomechanical testing. The results showed that replacing 20% to 30% of the tailings with aggregates in shotcrete production achieved compressive strengths exceeding 30 MPa after 28 days of curing, with an energy absorption capacity of 500 joules. Results show that this shotcrete can be used in the support stage of the mining cycle in underground mining.

Feb 22, 2026

By Alper Kirmaci, Taghi Sherizadeh, Dogukan Guner, Cameron Mitchel

Brow hazards are a significant concern in coal mine ribs and form when weak coal sloughs out of a rib face and exposes an overhanging block of more competent rock or coal. Of 36 rib failure fatalities in the last two decades, 36% resulted from rib brow failures. This highlights the need to better understand the mechanisms driving brow formations and failures to mitigate this hazard and eliminate potential fatalities. The research team at Missouri S&T worked to enhance the current CPRR system and include the potential for these brow hazards. We employed physical modeling methods to build a scaled model from critical brow cases observed in the field. The dimensions and properties of the models were scaled down by 14.4 times from the field cases. The results of the physical modeling efforts were used to assist in calibrating distinct element numerical models vital to understanding brow formation and failure mechanisms. Subsequently, three physical models were created and tested to provide data on an otherwise impossible-to-capture event at the field scale. The results of these tests provided formed brow dimensions that agree well with those observed at the field scale. The brow dimensions were 1.25 inches deep into the rib, equivalent to 1.5 feet when scaled up. Similarly, the numerical modeling efforts provided good agreement by capturing a brow forming 1.55 inches deep into the rib. The physical model results further validate the applicability of physical modeling efforts for capturing otherwise difficult and dangerous phenomena in a controlled lab setting.

Feb 22, 2026

By N. Sloan Dworian, Alyssa Smith, Ian Kellogg

Many studies have utilized magnetic susceptibility data to fingerprint porphyry copper centers; however, usage of magnetic susceptibility data to identify trends in a Climaxtype molybdenum porphyry deposit is limited. To better understand potential petrophysical signatures in a Climaxtype Mo porphyry deposit, magnetic susceptibility data was gathered at regular intervals from core holes across the Climax Mine, targeting major lithologies and styles of alteration. Preliminary analysis suggests that there are quantifiable patterns which may correspond with the mineralizing intrusions. This data can complement current mapping and exploration techniques to identify Climax-type molybdenum porphyry centers and likewise further the exploration potential for this deposit type.

Feb 22, 2026

By Ashish Kumar, Khadija Said, Ulupi Mishra, Sekhar Bhattacharyya

Using battery-powered equipment in underground mining can improve safety, advance automation, and reduce greenhouse emissions. In this context, rechargeable lithium- ion batteries (LIBs) have a competitive edge due to their high Coulombic efficiency, superior energy density, and extended cycling stability. However, several conditions must be met for their safe operation. Their failure can cause thermal runaway (TR), which poses a significant safety risk in underground mines. Despite studies having investigated this phenomenon using physics-based models, limited research has used deep learning to predict its implications. This study aims to bridge this gap by characterizing LIB TR under dynamic airflow through simulations and predicting temperatures using two of the most common recurrent neural networks: long short-term memory (LSTM) and gated recurrent units (GRU).

Feb 22, 2026

By Michael MacMillan

AI is already changing how mining companies are seen, judged, and trusted. This paper discusses “reputation engineering,” a structured way to build trust in a world where AI systems summarize and evaluate everything a company says publicly. Using mining examples, emerging AI search trends, and real communication work, the paper explains why the next 12–18 months are a crucial window for action. It also provides a practical framework to help mining companies improve their visibility, accuracy, and consistency across both human and machine channels.

Feb 22, 2026