Search Documents

By M J. Nimmo

extremely powerful tools for building predictive and generative models. ML can be used for building highly accurate regression and classification models. But without careful data science and statistical analysis, a regression or classification model could be highly biased and completely wrong – and we may not even realise. An example rock density estimation task will be used to illustrate the potential gains and possible pitfalls in using ML and AI in Mineral Resource estimation. What could possibly go wrong? Nothing! For most simple regression or classification problems, very little. For density estimation, using the global mean may be all that is possible, but the resource classification would need to reflect the uncertainty in global estimation of tonnage. Everything! From data collection and measurement, from selecting ML and AI techniques and designing the analysis workflow to our own cognitive bias. The things that could go wrong are vast and may include but not limited to asking the wrong question, not asking questions, data bias, measurement errors, insufficient data coverage, not collecting the right variables, not collecting enough observations, collecting clustered samples, unbalanced data, filtering out data without reason (outliers), clipping data (outliers), relying on automatic feature selection, including irrelevant variables, excluding relevant variables, geological interpretation, changing context, blindly following best practice, splitting small data sets, subsetting large data sets, using the wrong tool for the task, insufficient budget to complete the analysis, not allowing enough time for testing and experimentation (the science), focusing too much on building the model (the engineering), not learning from the data, our assumptions, our preconceptions, and our skill. This paper will explore the question of how training regression models to predict rock density are affected by gaps in the data (missing observations and missing variables). What could possibly go wrong?

May 24, 2023

By A M. Nyembwe, A M. Garbers-Craig, R Cromarty

With the depletion of high-grade lump iron ore, concentrates have been identified as an alternative material for sinter making. Adding concentrates to the sinter mixture reduces sinter bed porosity and gas flow through the sinter bed. This will result in lower sinter strand productivity. A potential solution to this problem is to add concentrate in the form of micropellets. This study investigates the effect of concentrate and micropellet additions on the permeability of the sinter mixture. Permeability was measured on sinter mixtures containing ten to 40 per cent concentrates and micropellets at various moisture contents. The effect of concentrate and micropellet addition on the permeability–moisture relationship was examined. The results confirmed that the addition of concentrates into the sinter mix decreases the maximum permeability of the sinter bed. The incorporation of micropellets into the sinter mix reduced the permeability less than concentrate addition. The study also showed that the maximum permeability of sinter mixes containing concentrate is reached at the same moisture content as the sinter mixture without concentrate addition. For sinter mixes containing micropellets, additional water must be added to obtain maximum permeability. Parameters including the granule size, void fraction and angle of repose were also investigated to evaluate their effects on the sinter bed permeability.CITATION:Nyembwe, A M, Cromarty, R and Garbers-Craig, A M, 2015. Effect of concentrate and micropellet addition on iron ore sinter bed permeability, in Proceedings Iron Ore 2015, pp 69–76 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Jul 13, 2015

By N Pranoto, E Lawell, T Dean, M Grant

The Goonyella Riverside Mine is located 30 km north of Moranbah in Central Queensland, Australia. Recent mining activities have exposed the north–south striking Kenya and Zimbabwe thrust Faults. These faults vertically offset coal seams by up to 25 m and locally thicken coal packages through imbrication of coal seams. The presence of back-thrusts add to the geotechnical complexity of maintaining highwall stability. The Kenya and Zimbabwe Faults have provided multiple challenges to the mining operation, with an unplanned additional 240 kt of coal tonnes being mined from imbrication of coals seams or from the refinement to local scale geological models. These additional coal tonnes have been either identified during the mining of the coal seam or just prior to overburden removal above the coal seam. These faults can also negatively impacted highwall stability with highwall failures, lowering the overall slope angle for pit designs and increasing stand-off distances for hazardous highwall conditions. Unmanned Aerial Vehicles are used to scan exposed highwalls and enable geological mapping via high-quality 3D images. Measure while drill data from drill and blast rigs can provide local-scale intersections of the faults and coal seams. Mine Geologists utilise these and other data sets to create localised geological models refined position of the fault planes and coal seams. These models are utilised by relevant mine site teams to ensure the safe and efficient mining of coal around these faults.

Mar 22, 2022

By B Muller

The disposal of tailings has become a greater issue due to the changing economics of mineral extraction. The processing of lower grade deposits and tougher environmental regulations has led to the selection process for tailings disposal options becoming more complicated. Current selection methods are too time consuming or too easily manipulated for inexperienced users to use successfully. This paper examines a different method for the initial selection of a tailings disposal system, using a numerical system that was introduced into the mining method selection field. The numerical method system was compiled by reviewing previous work done on the positive and negative aspects of the tailings disposal methods used by the coal and metallurgical mining industries. Assessing the common trends and design aspects of each of the methods was undertaken and a ranking was designed to explain how well suited the tailings disposal methods are to a situation. The tool successfully analysed sites using common data to indicate methods for disposal that would have provided minimal technical problems, low environmental risk and been practically possible. The tool also proved successful in identifying critical areas that would need further investigation. This new system was tested on some disaster case studies from around the world in order to show that it could have successfully selected a different method for tailings disposal that would have prevented the disaster.

Sep 26, 2011

By J Hadjigeorgiou, T R. Stacey

A ground support system is a combination of individual reinforcement and surface support elements that work together to retain and contain the rock mass. The various elements are subjected to external loads and to interactive loading between individual ground support system components. This paper focuses on the challenges in determining the capacity and behaviour of mesh, recognizing that installed mesh can be subjected to several loading mechanisms. Field observations from underground hard rock mines demonstrate that there are multiple types of damage and modes of failure of mesh. However, these are difficult to reproduce in a laboratory environment. A critical review has shown that existing mesh laboratory testing rigs reproduce very limited loading conditions that may not be representative of underground conditions. This has significant implications in quantifying and extrapolating the performance of mesh to ground conditions. CITATION:Hadjigeorgiou, J and Stacey, T R, 2018. Challenges in determining the capacity of mesh, in Proceedings The Fourth Australasian Ground Control in Mining Conference (AusRock), pp 280–290 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Nov 30, 2018

pNumerical modelling has been used to examine the relationship between the results of two commonly used methods of assessing the propensity of coal to spontaneous combustion, the R70 and Relative Ignition Temperature tests, and the likely behaviour in situ. The criticality of various parameters has been examined and a method of utilising critical self-heating parameters has been developed. This study shows that on their own, the laboratory test results do not provide a reliable guide to in situ behaviour but can be used in combination to considerably increase the ability to predict spontaneous combustion behaviour./p

Jan 1, 2005

The Broken Hill orebody ,occurs in the Willyama Complex, a block of Precambrian rocks which have been strongly folded and metamorphosed at moderate to high grade. Although -the origin of the silver-lead-zinc mineralization remains obscure, recent investigations have substantiated the view that the orebody has undergone concomitant high grade metamorphism with the enclosing wallrocks. The mesoscopic structures produced by this metamorphism are described, and the possible processes Telating to their formation are disoussed.In the high grade ore, metamorphic structures which can be recognized are foliation and lineation, wallrock fragments, layering, flow folds, ore breccias, and sulphide bearing veins. These structures can be related to processes involving plastic flow and fluid phase transport of the mineral assemblage. 'In areas of low grade mineralization adjacent to high grade ore, the structures more closely resemble those produced in -the wallrocks-foliatlon, lineation, layering, folds, and sulphide injections. Within areas of high grade ore, lenses and pods of manganese silicates Probably represent original inhomogeneities in the ore, i.e., portions of manganese-rich and sulphide deficient rock which have been metamorphosed to manganese silicates.From the structures displayed, a sequence of events has been established for the plastic flow of ore during deformation.

Jan 1, 1975

On the Kinetics and Equilibrium Models for Biosorption of Ni(II) onto a Gram-Positive Bacteria Strain, Rhodococcus opacus, for Environmental Applications The aim of this fundamental study was to evaluate the biosorption of Ni(II) from aqueous solution streams onto a Rhodococcus opacus strain through kinetics and equilibrium isotherms studies. The biosorption of nickel achieved its highest value in a pH around 5. The experimental data evaluation was carried out focusing three two-parameter models, as: Langmuir, Freundlich, Dubinin - Radushkevich. In order to determine the best isotherm fit, three error analysis methods were used to assess the data: correlation coefficient (R2), residual root mean square error (RMSE) and Chi-square test. The error analysis established that the Freundlich model fits better the nickel biosorption data. Maximum uptake capacity of nickel reached a value around 7.63 mg.g-1 for R. opacus. The biosorption rate was fast and achieved a removal of around 79 per cent of Ni(II) in the first 5 minutes at room temperature (25 ¦ 2¦C). The experimental kinetics data was evaluated considering two models (pseudo-second order and pseudo-first order). The error analysis established that the pseudo-second order model better fits the data.

Sep 13, 2010

By G Nicholas, S Coward, J Vann, R Wolff, S Jackson, S Moayer, A Bye

"A new approach to mineral project evaluation is described called Scenario-Based Project Evaluation (SBPE). SBPE incorporates well established æscenario planningÆ and æscenario thinkingÆ approaches, which are used extensively in other industries,?most notably the?petroleum industry. While most scenario planning is qualitative and non-probabilistic, the approach described here includes both quantitative and qualitative aspects. SBPE is complementary to optimisation approaches and allows mining firms to build and operate projects that are more robust in the face of adverse changes as well as proactively respond and exploit the upside offered by a realistic assessment of variability. The approach has potential to improve project performance and decrease operational risk. It also has important applications for currently operating mines.Conventional evaluation of metalliferous mineral deposits uses estimates of metal grades and deleterious attributes in singular æorebody modelsÆ, usually 3D block models, which fail to capture two critical aspects that underpin the realised value of mining projects: the inherent uncertainty of the estimated attributes the spatial variability of attributes, at the scale of mining selection, which impact on downstream processing steps and fundamentally drive realised value. While singular models are appropriate for estimating block averages, conditional simulation (CS) models of the relevant attributes are needed to represent uncertainty and capture spatial variability. Any attributes that can be either built into CS models and/or modelled in the envisaged mining/processing steps, can be evaluated fully through to highly ægranularÆ (ie detailed and un-smoothed) cash flows. A complete view of SBPE relies on selected strategies being tested against plausible æscenariosÆ of factors external to the project and thus outside management control. æExternalitiesÆ captured in such scenarios include market prices, costs and taxes. Via SBPE, project managers may demonstrate that there is material value in preserving or purchasing options that provide future flexibility to respond to the changing externalities or changes in orebody performance. Such flexibility generally comes at a cost but is very difficult to value. It is argued that full value chain evaluation using SBPE can help to value these options for a mining company. A case study for a gold mine is presented.CITATION:Vann, J, Jackson, S, Bye, A, Coward, S, Moayer, S, Nicholas, G and Wolff, R, 2012. Scenario thinking - A powerful tool for strategic planning and evaluation of mining projects and operations, in Proceedings Project Evaluation 2012, pp 5-14 (The Australasian Institute of Mining and Metallurgy: Melbourne)."

May 24, 2012

The paper continues an account, begun in a previous paper, of experimental work carried out by the author whilst a member of the metallurgical staff at the Cockle Creek Works. The general method of the description of the plant and works practice adopted previously has been followed.The paper is a short account of the methods used in and the results of an investigation of the fume losses from the various parts of the plant.MEASUREMENT OF VOLUMETwo independent methods were used. The first depended on a number of pitot-tube measurements of velocity. These readings were taken in the stacks leading from the various furnaces at a considerable height above the ground so as to obtain as regular conditions as possible at the points where the readings were taken.Three openings were made in each stack and the pitot tube inserted into each opening in turn and readings taken at the centres of imaginary areas of equal size. The average velocity in the stack was thus obtained.After four tests had been made at the smelter stack, which removed the gases from all blast furnaces in use, the centre of velocity of the gases for each test was calculated and was found to be practically constant in position. The previous procedure was varied in subsequent tests on this stack by taking a succession of readings at the centre of velocity instead of at various points...

Jan 1, 1925

It is possible to divide industrial minerals into four general categories: 1. fuels, 2. construction materials, 4. agricultural minerals, and 4. industrial minerals. It is proposed to ignore the fuels in the context of this paper and deal in general terms with the others. Mining companies have often considered getting into the industrial mineral business because they can see a resource development with good growth and consistent profits. However, many mining companies being conversant with the metallic side of the industry have spent sleepless nights determining which industrial minerals are most suited for their particular situation. Some have got into the business for the wrong reasons and wished they had never heard of it. Others have successfully adapted to the industrial minerals philosophy and now operate in both metallic and industrial mineral arenas. There is a large difference between firstly, the philosophy and secondly, the functions of being in the industrial minerals business as opposed to metallicÆs. The philosophies can be considered in parallel but it is necessary to clearly have an appreciation of the dominant influences that affect industrial minerals.

Jan 1, 1984

The Ediacara mineral field is classified as a Mississippi Valley type deposit with stratiform mineralization near the base of the Cambrian beds. Mineral deposition was contemporaneous with sedmentation and reerystallization, and mobilization took place after lithification of the sediments. Several mechanisms of deposition were possible.INTRODUCTIONThe Ediacara mineral field is located in the Mt. James Range approximately 300 miles north of Adelaide, about 12 miles west of Beltana railway siding and 10 miles east of Lake Torrens. It is isolated geographically from similar deposits of equivalent age and environment in the Flinders Range Province (Fig. 1). The climate may be described as semi arid with high summer temperatures and frequent dust storms. Occasional thunder storms are the main source of rain in the summer months.The exact date of discovery is unknown, but is believed to be about middle of the last century. Ulrich (1872) was the first geologist to report on the field. At that time only the southern portion of the field had been prospected (and abandoned). Although the prospects did not appear encouraging Ulrich advised prospecting of the shales and testing beneath the limestone.In 1888, Rosewarne, an Inspector of Mines, visited the Southern Workings and queried whether they were connected with some lode at deeper level or were merely bedded deposits. The first geological map with sections was drawn by H.Y.L. Brown (Brown 1892).Brown recognized three principal rock formations as dolomitic sandstone, kaolinised shales and sandstone, and quartzites and sandstone in descending order. These three basic units are still recognized. He recognized also the stratiform nature of the mineralization in the shales as a...

Jan 1, 1963

How often does one hear that a recent feasibility study was not a success? Alternatively, when scope of the feasibility study was sent out to tender there was a large discrepancy in tendered prices. If the tendered prices were to be compared to the consumer market, for instance the purchase of a car, where the concept is transport in a controlled and safe manner, prices vary by large factors, but then so do the specifications. In the case of a feasibility study, prices should be within a tighter margin as the tenderers are all quoting against the same concept and specification. Why do these discrepancies occur? Price discrepancies will in every likelihood be attributed to two main factors; one, the perceived amount of work and two, the ability of the bidder to carry the risk associated with the feasibility study. It is, therefore, important that the scoping of the feasibility study covers the detailed requirements as clearly as possible stating what level of accuracy is required. The feasibility study is one of the first activities in the life of a project. The study, like the project it is investigating, is subject to the constraints of time, cost and quality. Its primary objectives are to demonstrate the viability of a project within the outlined scope and constraints.

Jan 1, 1998

By A T. Job, P R. McAree

There is significant value at stake in the mining industry from the implementation of new technologies that support the enhanced profitability and sustainability of mines. However, in order to realise this value, new technology must be implemented in a way that is both effective and sustainable. The industry has a mixed history for technology implementation that provides fertile ground for understanding what is necessary for innovation to deliver value. This paper describes early stage research investigating why this is so, with the aim of improving how to deliver value. Several case studies are provided covering both successful and unsuccessful technology interventions. The paper identifies the common factors that emerge from the case studies, and offers a path forward for further research in this area.CITATION:Job, A T and McAree, P R, 2017. Three case studies on the implementation of new technology in the mining industry, in Proceedings Iron Ore 2017, pp 407–414 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Jul 24, 2017

COMPRESSED air is held in high esteem by the mining engineer as a medium for the transmission of power. The transmission, however, is accompanied by a high percentage of loss. Of the total work done by the steam cylinder'in compressing the air inthe air cylinder, less than 50% is ultimately utilized by the machines underground. These losses are due to its essential and inherent properties as a gas. One of the sources of loss is that due to transmission through the pipes which distribute the air from the receiver to the various places where it is used. As the capacity and economic design of air-compressing plant are influenced by this source of loss, it may be of interest to give an outline of how estimates may readily be made'in practice of thediameter of pipes, quantity of air discharged, and the drop in pressure occasioned.In hydraulics we have a general formula given M Rankine's "Civil Engineering" :It was suggested, the writer believes, in the first instance by D'Arcy, that this formula could be made to apply to an elastic...

Jan 1, 1918

It is very rare in the underground hard rock environment to find a deposit that is ideally suited to a specific mining method. Usually the irregular shape of the orebody will allow the deposit to be extracted by more than one mining method. During the prefeasibility stage, a number of conceptual mine designs are generally evaluated to see which offers the best method of extraction for the deposit. The selected method is usually based on key decision criteria such as the overall extraction of the orebody, the amount of metal produced annually and the unit cost of production. Often the prefeasibility study document will state that the project has an NPV of æxÆ million dollars. This statement can be made without any quantitative consideration of the likely risks of the mining methods reviewed or the level of confidence in the estimation of the NPV. By assigning a probability of failure to each component of the mining cycle a measure of the risk for each method can be determined. One mining method may provide the highest NPV when it is performing well, however, there may be a higher risk of a major failure impacting on the performance of that method, eg the failure of a mass blast of a crown or rib pillar. This paper will present a flowsheet approach to the use of risk analysis for determining the likely production rates of different mining methods. This framework allows the annual metal output to be estimated and from this the NPV for each method. A level of confidence can be applied to the estimated NPV for each option. An example will be presented on how this technique has been used within the Normandy Group.

Jan 1, 1998

By T Hughes

The Gekko InLine Pressure Jig (IPJ) was unleashed on the World gravity separation market in 1996 and has since sold over 150 units in applications including tin, diamonds, silver, coal, garnet and gold. During the past twelve years the knowledgebase of the best applications and capabilities of the IPJ has greatly increased. Improvements to the original IPJ design have included engineered ceramic ragging in specific gravities from 1.6 to 5.0, full automation and improved designs of feed presentation and tailings and concentrate handling systems. The IPJ has been found to be able to handle a wide feed size range and recover mineral at liberation size without significant addition of water to the mass balance. Test work carried out during full-scale operation of the IPJ has shown the IPJ can recover diamonds down to 0.5 mm, sulfides down to 0.3 mm and gold down to 0.1 mm which greatly expands its application in the mineral processing industry. A significant volume of data now supports the symbiotic benefits of an IPJ and flotation circuit combination to pre-concentrate an ore and minimise downstream processing requirements.

Jan 1, 2008

By S Sullivan

A recent review of resource reports lodged to statutory bodies found geological models with oversmoothed or unrealistic lithological boundaries, many of which do not represent geological appearance in the field, pit or underground drive. The underlying data is often too complex or messy to be incorporated into the resource model and is thus a simplification of reality, to the detriment of the mining operation and its shareholders. Volumes can be over- or underestimated and subsequent grade predictions can be misplaced, combining to cause over- or understatement in resource reporting to stakeholders and statutory authorities. Domain models can be improved with new modelling techniques such as machine learning. However, with poor data or improper use of machine learning techniques, the conclusions can be just as misleading as with current techniques. Machine learning rarely offers a single ‘right’ outcome, rather a range of possible outcomes from the data provided. But importantly it can take advantage of the full richness of data for more informed models. Machine learning also provides confidence in its predictions so that uncertainty in the resultant models can be quantified and passed through for risk assessment during mine planning and operation. This paper will discuss the factors accentuating complexity in deposit modelling: data diversity, structural controls, chemistry, data volumes, process workflows and external non-geological constraints. A case study illustrates the risk of reducing or ignoring complexity, which can result in an overly simplified geological model.

Mar 22, 2022

By A Atkins

Six recent case studies were selected based on the eight main causes of mine disasters (Table 1; Quinlan, 2014) and where credible information sources, such as Royal Commission Reports, Coronial Investigations and textbooks, were available. The six case studies are listed below. Details are available via this link or the conference PowerPoint accompanying this paper. 1. Beaconsfield, Tasmania – 2006 Fall of Ground/Seismic Event (Chandler, 2009; Melick, 2007; Wright, 2012). 2. Bingham Canyon, USA – 2013 Manefay Slide (Ross, 2017). 3. Bronzewing, WA – 2000 Hydraulic Fill Inrush (Hope, 2002). 4. Brumadinho and Samarco, Brazil – 2015 and 2019 Tailings Storage Facility (TSF) failures (Bowker, 2019; Darlington et al, 2019; Demajorovic, Lopes and Santiago, 2019; Morgenstern, 2016). 5. Northparkes, NSW – 1999 Cave Air-blast (Bailey, 2003; Hebblewhite, 2003). 6. Pike River, NZ – 2010 Methane Gas Explosion (Macfie, 2013; Panckhurst, Bell and Henry, 2012.

May 5, 2022

It has been fairly firmly established (Fisher, N. H., 1945) that the fineness of the gold in any ore deposit varies with the depth from the surface at which the deposit was formed, and, as a corollary to this, that it is dependent upon the temperature and pressure at the time of deposition. This relationship is such that, under certain conditions, the gold fineness, taken into consideration with other recognised criteria, furnishes a very sensitive and reliable guide to the relative temperature of ore formation, at least within the epithermal and the upper part of the mesothermal range.Definition of Fineness.The fineness of gold is calculated in parts per thousand and refers to the proportion of gold present in the naturally occurring metal in the lode, which is actually a gold-silver alloy, containing usually about 10 parts per thousand of base metal and up to 500 parts perthousand of silver. A clear distinction should be made between fineness and gold-silver ratio. The latter refers to the total gold and silver in the lode as determined by assay and is usually expressed as the ratio Au:Ag. Only in deposits where other silver-bearingminerals are entirely absent, which occurs comparatively rarely, can the gold-silver ratio be used as a measure of gold fineness and, in many cases, the two have no relationship. For instance, in the Edie Creek area in New Guinea, the lodes formed under shallow epithermal conditions all carry gold 500 to 550 fine, although the gold-silver ratio ranges from 1:100 to 1:1, in places in adjacent lodes.Determination of Gold Fineness.The fineness of lode gold can be determined only after it has been separated from all other minerals. Amalgamation recovers gold fairly cleanly - unless native silver is present - and the fineness of amalgam bullion is generally a reliable indication of the correct fineness of the gold. If the use of plates is involved, the bullion may be adulterated, but the effect of any copper, or other metal, that may have been introduced can be cancelled by assuming that the gold and silver shouJrl add up to about 990 fine, or by ignoring the base metal content altogether, which varies little in naturally-occurring gold, and in calculating for comparison purposes all fineness figures to the basis Au + Ag = 1,000.The composition of bullion obtained from cyanidation is generally intermediate between the true gold fineness and the total gold-silver proportion as determined by assay of mine or mill-head samples of Cyanide solution dissolves other silver-bearing minerals and the fineness of the resulting bullion is thereby lowered. Consequently, the fineness of cyanide bullion rarely corresponds to the true gold fineness and is useless for diagnostic purposes, although it is a common fallacy to refer to the fineness of cyanide bullion as the fineness of the gold.

Jan 1, 1950