Search Documents

"The Archaean komatiite-associated nickel sulphide deposits at Kambalda, Western Australia, exhibit a number of stratigraphic features that are indicative of a strong volcanic control on ore localization, including (1) a restriction to specific basal cumulate metakomatiite host units, (2) confinement within distinctive synvolcanic embayments in the footwall and (3) variations in the internal structure and composition of overlying and adjacent barren metakomatiite flow's and the distribution of interflow metasediments with proximity to the ore zones.The high magnesium content of the host units (up to 45% MgO volatile-free) in comparison with that of aphyric or spinifex- textured lavas (16-32% MgO) has been considered previously to indicate either magmas/lavas rich in intratelluric olivine or selective accumulation of olivine phenocrysts. The marked linear trend of the oreshoots, basal host units, footwall embayments and the volcanic stratigraphic relationships have been used as evidence for local linear feeding fissures. Rare relict igneous olivine in some host units is, however, demonstrably not intratelluric, and there is no direct or indirect evidence of a proximal eruptive site (e.g. footwall feeders, pyroclastic breccias, ultramafic intrusives) at Kambalda such as at some other deposits of this type. Kambalda is interpreted as a distal rather than proximal volcanic environment.Emplacement of the komatiite sequence at Kambalda was probably controlled by the locations of a few specific eruptive sites relative to regional topographic gradients and influenced by the topography of the pre-existing volcanic surface. Olivine enrichment in the basal host units occurred by in-situ fractionation and cumulus crystallization of olivine during longitudinal lava flow within linear lava conduits, possibly centralized on depressions between non-overlapping footwall basalt flows. The distinctive stratigraphic relationships are attributed to localization of lava conduits along the volcanic pitchline, flanked by more slowly moving parts of the flows. There are analogies with the physical volcanology of Modern basaltic and Proterozoic komatiitic lava flows.A corollary of this model is that the lava conduits thermally eroded and assimilated sulphidic sediments, contributing additional sulphur to sulphur-saturated komatiites or possibly inducing sulphide separation in originally sulphur-undersaturated komatiites. The latter would account for the restriction of ores to linear cumulate komatiite host units (thermally active lava conduits) and the locally mutually exclusive, but stratigraphically equivalent, relationship between the ores and sulphidic metasediments at Kambalda."

Jan 1, 1984

By Valois R. Shea-Albin

Ground control problems are an important factor affecting safety, production, and efficiency in underground coal mines. A major portion of ground control problems encountered in underground coal mines can be attributed to geologic features in the strata surrounding the extracted coal seam. The U.S. Bureau of Mines has compiled information from numerous sources on the geological features that contribute to ground control problems in underground coal mines. The compilation includes (1) sedimentary features such as paleochannels, crevasse splays, clastic dikes, mold-and-cast structures, concretions, and lithologic factors, and (2) structural features such as folds, fractures, joints, cleat, and faults. The compiled information will aid in identifying the features, predicting their occurrence in advance of mining, and controlling or minimizing roof failure when these features are encountered in an underground coal mine.

Jan 1, 1993

By I. H. Douglas

All grade control methods in use currently by the mining industry are imprecise. The methods suffer from being locally inaccurate, leading to are loss and contact dilution. An ore/waste indicator variable can be used to predict the probability of ore occurrence at a given location, Po. Then, an ore grade Zo for an ore location and a waste grade Zw for a waste location are obtained. The expected revenue is calculated at each location, based on its probability of being are or waste, the corresponding grade, and on the cost of mining and processing the ore or the waste. Finally, the decision of whether to recover the block is based on the expected revenue value for that block. The method is illustrated with an example taken from the Jerritt Canyon district in northeastern Nevada.

Jan 1, 1994

The panic bars (and associated linkages and electrical circuitry) that are defined in this document are a machine operator-actuated control for the emergency deenergization of the tram motors on self-propelled, electric and diesel face equipment; including haulage equipment such as shuttle cars. There are nine basic types of panic bars (and several variations) that are recommended in this document, but all of the panic bars are characteristically of the ?fire door? type that is frequently encountered on outside exit doors of many public and some commercial buildings. It is also the intent of the following panic bar designs that, when emergency brakes are added to the face equipment machinery for which they are required, that they be designed and hooked up so that actuation of the panic bar will result in automatic application of the emergency brakes (immediately following the deenergization of the tram motors). The following panic bar design requirements are a restatement of the proposed requirements defined in the Federal Register, requirements cited in "Guidelines for the Design and Installation of Devices for Deenergization of Self -Propelled Electric Face Equipment/Automatic Emergency Brake s II by C. E. Lester and J. D. Greenhalgh of MESA, and Bendix recommendations that were formulated as a result of discussions with personnel at the various machine manufacturers and underground coal mines that we surveyed, discussions with G. L. Judy (our consultant), etc:

Jan 1, 1974

By Jue Kou, Lei Yu, Bern Klein, Lusheng Zhou, Chunbao Sun, Genzhuang Li, Reem Roufail, Amit Kumar

Magnetic liners can improve the performance of ball mills by increasing grinding efficiency, extending service life, and reducing maintenance costs. Despite their importance, the fundamental understanding and the quantitative investigation on the effects of the magnetic liners has not been addressed in the literature. This paper addresses this gap by studying charge motion, and how the use of a magnetic liner affects the breakage mechanism of the mill, using discrete element method (DEM) modeling and the Hertz-Mindlin contact model with relative velocity dependent (RVD) rolling friction. Charge motion was modeled under different mill speeds. Compared to conventional steel and rubber liners, the charge motion in the magnetic liner mill was predominantly cascading. Quantitative analysis of the energy dissipation within the mill was conducted to investigate mill breakage. Regardless of speed, over 50% of the total energy dissipation in the magnetic liner mill was in the form of abrasion and attrition. The results highlighted the appeal of magnetic liners in secondary and regrinding application, where abrasion and attrition are considered more efficient in the fine material breakage, and where impact is ideally minimized to reduce mill wear. A number of experiments were conducted to assess the effect of different model parameters on the performance of the magnetic liner mill. The highest abrasion/attrition intensity was observed with high values of the model parameters namely the restitution coefficient, static friction coefficient, and rolling friction coefficient.

By N. A. Gjostein

The formal descnption of the decay of an isolated scratch can be written in terms of an appropriate Fourier integral. With the application of certain approximations, this description leads to the second-moment analysis developed by King and Mullins. It is shown that the second-moment analysis does not adequately describe the behavior of a decaying, isolated scratch, and that it is difficult to obtain internally consistent values of the surface self-diffusion coefficient, D,,from this analysis. This inconsistency arises from contributions of higher-order moments to the decay process. Two new analytical methods, based on the Fourier integval representation of a scratch profile, have been developed for determining values of D, from the decay kinetics of isolated scratches. In the profile-matching method, a given initial profile is resolved into its Fourier components, the components are allowed to decay exponentially over an annealing time interval, At, and finally they are synthesized to form a decayed profile. The predicted profile is adjusted to fit the experimental profile by varying only the parameter, D,At; the contribution from volume diffusion is directly taken into account and the integrals involved in the Fourier analysis are evaluated by numerical integration on an electronic computer. In the kinetic method, the widths, W, of a series of predicted profiles are determined for various volume-diffusion contributions, and the results are plotted in the form W vs DSAt. For a fixed ratio of surface to volume diffusivities, R, these plots are linear; the slope of the lines increases as R decreases. A simi-lar analysis can be performed on the depths of the predicted profiles. In this case, plots of (l/df3 vs D,At are found to be nonlinear. The proper value of Ds is found by superimposing the data for the experimental widths and depths on Plots of this kind and adjusting Ds until a fit is obtained with the line having the appropriate value oj R. An analysis of the slopes of these plots for R —= (surface diffusion only) reveals that they do not give the correct value, 2276, for width tf second-moment standard profile. This finding allows one to modify the second-moment analysis in such a way as to obtain internally consistent values of Ds. ThERE are several processes, involving a flow of matter along a surface, due to gradients in curvature of the surface, that may be used to determine the surface self-diffusion coefficient, Ds. These processes are commonly known as 1) thermal grooving, 2) single and 3) multiple scratch decay, 4) blunting of field-emitter tips, and 5) sintering. In this series of papers, the first three of these processes are examined in detail in an attempt to subject the phenomenological theories governing these processes to a critical test and, furthermore, to make a critical comparison of the values of D, obtained from these techniques. Although the thermal grooving technique has been established'' as a reliable method for determining D, , this has not been the case for the process of scratch flattening. Mullins and coworkers374 have proposed phenomenological laws governing the growth of the width and the decay of the amplitude of an isolated scratch, but only in one study5 has the width-growth law been supported by experimental evidence. In paper I of this series detailed experimental evidence is presented that points up the deficiencies of the second-moment analysis; the origins of these deficiencies are also discussed. Following this development, some new analytical procedures, based on representing the scratch profile by Fourier integrals, are presented as a means of obtaining reliable values of D, from the kinetics of the scratch-flattening process. Further applications of the Fourier integral approach are reported in papers 11' and 111,~ where the method has been useful in interpreting the anisotropy of flattening rates of isolated scratches (paper 11) and in describing the multiple scratch decay process (paper 111). THEORY OF SCRATCH DECAY Following ~ullins" approach we consider an arbitrarily shaped scratch profile, Fig. 6, at some time, t. The profile is designated by a function z(x,t) which can be expressed in terms of a Fourier integral as follows:

Jan 1, 1967

By H. C. Fuller

A method was devised and developed by the Bureau of Mines for decomposing manganese sulfate at significantly lower temperature than that required by the conventional procedure of using heat alone. This technique, called the partial reduction method, effected decomposition at 700° to 800° C, as compared to 900° 1,000° C normally required. The process consisted of heating, in the absence of oxygen, a mixture of manganese sulfate and a controlled quantity of a reducing agent. The suitability of commercial-type equipment was tested in laboratory-size models of an externally heated rotary kiln, a r bbled muffle-hearth furnace, and a fluidized-bed reactor. The rabbled-hearth furnace gave the best results with the least operating difficulty produced a gas containing more than 80 percent SC and eliminated 96 percent of the sulfur when coal was used as the reductant. Sintering of the powdery decomposition product produced an agglomerate containing, in percent, 63.6 Mn and 0.4 S.

Jan 1, 1966

By J Lombardi

OZ Minerals' Prominent Hill copper-gold concentrator is located 130 kilometres south east of the town of Coober Pedy in the Gawler Craton of South Australia. The concentrator was built in 2008 and commenced commercial production in early 2009. The Prominent Hill concentrator is comprised of a conventional grinding and flotation processing plant with a 9.6Mtpa ore throughput capacity. The flotation circuit includes six rougher cells, an IsaMill for regrinding the rougher concentrate and a Jameson cell heading up the three stage conventional cell cleaner circuit. In total there are four level controllers in the rougher train and ten level controllers in the cleaning circuit for eighteen cells.Generic proportional – integral and derivative (PID) control used on the level controllers alone propagated any disturbances downstream in the circuit that were generated from the grinding circuit, hoppers, between cells and interconnected banks of cells, having a negative impact on plant performance. To better control such disturbances, FloatStar level stabiliser was selected for installation on the flotation circuit to account for the interaction between the cells. Multivariable control was also installed on the five concentrate hoppers to maintain consistent feed to the cells and to the IsaMill. An additional area identified for optimisation in the flotation circuit was the mass pull rate from the rougher cells. FloatStar flow optimiser was selected to be installed subsequent to the FloatStar level stabiliser. This allowed for a unified, consistent and optimal approach to running the rougher circuit. This paper describes the improvement in the stabilisation of the circuit achieved by the FloatStar level stabiliser by using the interaction matrix between cell level controllers and the results and benefits of implementing the FloatStar flow optimiser on the rougher train.

Aug 8, 2011

By T. F. Anstett

The Bureau of Mines estimated the potential availability of rare-earth oxides (EtEO), including yttrium, and thorium, which is also contained in the rare-earth bearing minerals monazite and bastnasite, from 38 'properties in market economy countries (MEC's). Only nine of the properties evaluated produce or would produce REO as the primary product; the others contain REO as a byproduct, chiefly from mineral sands operations recovering rutile, ilmenite, and zircon. Nearly 77% (2,578,000 mt) of the total 3,355,000 mt of recoverable RE0 evaluated is from producing properties. It is estimated that 21% (705,000 mt) of this total is from properties that could not realize a positive rate of return at the January 1984 market prices for the recovered commodities. At production capacities assumed for this evaluation, the amount of RE0 potentially available from producers ranges from an estimated 46,500 mt in 1986, to a peak of 50,900 mt in 1992, to 49,900 mt by the year 2000. Total MEG production in 1984 was estimated at 40,700 mt. Assuming demand does not increase sharply, producing properties can continue to fulfill overall REO demand through at least the end of this century.

Jan 1, 1986

By R Sinclair, D Landwehr, I Hunter

"The Syama Gold Mine, located in southern Mali, was established in 1990 by Broken Hill Propriety Limited (BHP). The operation began treating oxide, and in 1994 it transitioned to processing hard and refractory ore based on ‘whole-of-ore’ roasting. In 1996, BHP was succeeded by Randgold Resources and the operation ran until 2001, when it was placed into care and maintenance.The Syama ore is classified as a double refractory ore with pyrite occluding the fine grained gold present in the ore and high levels of organic carbon presenting a challenge in terms of preg-robbing behaviour. The ore is extremely competent and this combination of metallurgical features was the main source of metallurgical problems encountered by the previous owners.Resolute Mining Limited (RML) acquired the Syama project in 2004. Following a feasibility study and the development of a revised processing flow sheet the plant was recommissioned in 2008 using a combination of old and new equipment. The revised flow sheet enabled primary sulfide ore to be treated by flotation and roasting of the concentrate to produce a calcine suitable for conventional carbon-in-leach (CIL) processing.Commissioning and ramp up of the plant was a troubled process, with monthly gold production being as low as 3000 oz in 2010. Poor availability of equipment and lower-than-forecast recoveries were the primary contributors.During the last five years, RML has utilised a systematic approach to improve production, focusing on mechanical reliability, metallurgical recovery and process bottlenecks. Modifications to the circuit have been as a result of both innovation and the application of technological improvements. Improvements have been made to the flotation, roasting and CIL circuits and RML has introduced a deslime circuit to treat flotation concentrate, in addition to the very successful commissioning of a new 1 Mt/a oxide circuit in December 2014. These changes have resulted in gold production increasing to more than 20 000 oz per month in 2015.This paper presents the processes, change implementation and outcomes of the modifications made to the sulfide circuit to address the availability and recovery problems.Apart from the expansion of an existing, world-class deposit and discovery of new resources, future improvements on the path to a class-leading gold mine include the continued reduction of the unit cash cost, increase in overall sulfide circuit gold recovery, increased throughput through continued innovation and consistency in the plant’s operation.Citation:Sinclair, R, Hunter, I and Landwehr, D, 2016. Syama Gold Mine - the Journey towards a Class-leading Gold Mine, in Proceedings 13th AusIMM Mill Operators’ Conference 2016, pp 213–224 (The Australasian Institute of Mining and Metallurgy: Melbourne)."

Oct 10, 2016

By William P. Diamond

This report presents results of a cooperative research project between the U.S. Bureau of Mines and BethEnergy Mines, Inc. Four coreholes (two before mining and two after mining) were drilled at a longwall mine operating in the Lower Kittanning Coalbed to obtain coal and rock samples from over-lying strata to determine their gas content at various times in the mining cycle. The results of those tests indicate that 91 pct of the gas removed from the overlying strata came from coalbeds. Material balance calculations were made to compare the volume of gas produced both from gob gas vent holes drilled on the panel and gas removed by the mine's ventilation system with the volume of gas removed from strata directly overlying the panel to a height of 275 ft. This analysis indicates that only 40 pct of the total gas produced from the panel actually came from the strata directly overlying the panel. The remaining volume of gas production probably migrated to the longwall gob from overlying, and perhaps underlying, strata immediately adjacent to the panel and perhaps from greater distances downdip because of the establishment of a long-term pressure gradient in the study area.

Jan 1, 2010

By H. Wotruba, L. von Ketelhodt, M. Kowalzcyk, M. Robben

ABSTRACT Decreasing reserves, water shortages, increasing energy costs, and environmental and social constraints are challenges to the gold mining industry that must be met with innovative machines and methods. This paper introduces the principles of sensor-based sorting and installations that demonstrate the technical and financial viability of the method. The scenarios discussed highlight the potential and advantages of this dry processing technology that requires little infrastructure. Case studies are presented for different ore types using optical- and X-ray transmission–based sorting. Current research is discussed in terms of the potential of near-infrared spectroscopy and X-ray fluorescence sorting for gold applications.

Jan 1, 2013

By S. Klein, M. McLeod, S. Gambino, K. Johnson, E. O’Hara

The Mokelumne Aqueducts Resiliency Project (MARP) involves construction of a 16.5-mile tunnel across the Sacramento-San Joaquin Delta (Delta) to replace the existing surface aqueduct pipelines with more resilient conveyance. The tunnel will be bored in Pleistocene-age alluvium at elevations between 100 to 10 feet below mean sea level. At this depth ground conditions are signicantly stiffer and denser than surcial soils and will mitigate potential seismic harards related to liquefaction ground shaking and cooding that could arise from potentially damaged levees. The heterogeneous soil deposits at tunnel depth consist of interbedded layers of stiff to very stiff clay and silt, and dense to very dense clean sand and silty sand.

Jun 23, 2024

By L. E. Mercando

The proposed Flambeau Mining Project is based on a high grade copper deposit located approximately one mile southwest of Ladysmith, Wisconsin. The project involves only the mining, crushing, and shipping of ore to facilities outside of Wisconsin for processing. The Wisconsin environmental regulations include provisions for identifying wetlands, the requirements for minimizing the disturbance of wetlands, and the need to establish new wet-lands to replace any wetlands which have to be disturbed.

Jan 1, 1991

By A. R. Chambers

Malagash received its name from the Indians many years ago on acco.unt of its turbid (milky) waters, although,' the red men did not appreciate the significance of this turbidity, and were, of course, entirely ignorant of the crystal beds of pure salt that lay beneath their feet. The salt lake of the Mississippian, or Lower Carboniferous period, in which these beds occur, was apparently a large one, for there is evidence of its existence at intervals all the way from Sydney, Cape Breton, to St. John, New Brunswick, 300 miles as the crow flies, and there is no knowing what its other dimensions may have been, as much of this horizon is now buried on the one side beneath the waters of the Gulf of St. Lawrence, and on the other underneath the Atlantic ocean. About midway of the area and underlying the counties of Colchester and Cumberland, there occurs a synclinal basin of which the Malagash salt outcrop forms only a small portion of its northern limit. This basin has been described very fully by A. C. Hayes. (1) The operations carried on at Malagash since Hayes wrote his report have brought to light very little in the way of new geological data, except to confirm his pre-dictions as to the probable faulting of the salt strata, and the effect which these faults would have on mining operations.

Jan 1, 1924

By Dwight D. Trueblood

The Benthic Impact Experiment (BIE) is designed to address the effects of sediment redeposition prior to the commencement of commercial mining. The experiment is being conducted in collaboration with Russian scientists from the Yuzhmorgeologiya Association aboard the R/V YUZHMORGEOLOGIYA. Sediment resuspension and subsequent redeposition during manganese-nodule mining is predicted to be one of the primary impacts on benthic communities living on the abyssal seafloor. The redeposition thicknesses that will cause significant faunal mortality in the deep sea are unknown. Sediment redeposition could adversely affect abyssal benthic communities through entombment, or through starvation caused by food dilution and obstructed feeding. This spring NOAA conducted the main portion of the BIE experiment, blanketing the experimental area with varying thicknesses of sediment using the Deep Sea Sediment Resuspension System (DSSRS). The DSSRS was conceived by NOAA scientists, designed and built by Williamson and Associates, and is the critical piece of equipment used to generate the controlled sediment redeposition environment on the 4800-m deep abyssal plane. The DSSRS consists of an 1800-kg sled, 4.8 m long, 2.4 m wide, and 2.2 m high. The sled was towed and powered using an 8,000-m, 26-mm single conductor cable having a 3,000 VAC capacity. As the sled was towed across the bottom, mud was ingested through two rear-facing, 45-cm wide intakes using two 7.6-horsepower pumps, and discharged 10 to 20 m above the bottom through two 20-cm diameter riser hoses. Each pump was instrumented with optical backscatter sensors providing real-time information about the discharge concentration.

Jan 1, 1992

By Yousef Ghorbani, Jafar Abdolahi Sharif, Fatemeh Kazemi, Ataallah Bahrami

Kinetic models are the most important tool for predicting and evaluating the performance of flotation circuits. Gilsonite is a natural fossil resource similar to an oil asphalt, high in asphaltenes. Here, in order to determine the kinetic order and flotation rate of a gilsonite sample, flotation experiments were carried out in both rougher and cleaner stages. Experiments were conducted using the combinations of oil–MIBC and gas oil–pine oil, with one test without collector and frother. Five kinetic models were applied to the data obtained from the flotation tests using MATLAB software. Statistical analysis showed that the results of the experiment with oil–MIBC were highly in compliance with all models. Kinetic constants (k) were calculated as 0.1548 (s−1 ) and 0.0450 (s−1 ) for rougher and cleaner stages, respectively. Rougher and cleaner tests without collector and frother also matched all models well (R2 > 0.98), with k values of 0.2163 (s−1 ) and 0.284 (s−1 ), respectively. The relationship between flotation rate constant, maximum combustible recovery, and particle size showed that the maximum flotation combustible recovery and flotation rate were obtained in the size range of −250 + 106 μm in the rougher and cleaner stages. The combustible recovery and flotation rate were higher in the rougher flotation process than in the cleaner stage.

By A. Vilinska

Microorganisms have a tremendous influence on their environment through the transfer of energy, charge, and materials across a complex biotic mineral-solution interface. Recent developments in biotechnology have not only aided hydrometallurgical leaching operations but also shown great promise in mineral beneficiation processes. The adhesion of bacteria on minerals alters the surface properties relevant to flotation and flocculation of minerals. The biomodification of mineral surfaces occurs through the attachment of microbial cells, oxidation reactions and adsorption or chemical reaction with the metabolite products. Several types of heterotrophic and chemolithotrophic bacteria have been successfully applied in flotation and flocculation processes for the last two decades. Al-though the potential of microorganisms in minerals beneficiation has been realised by several re-searchers from their laboratory investigations, the studies were mostly confined to pure mineral systems and there is hardly any sincere effort to apply the methodologies on ores. This paper presents a brief summary of the literature in bioflotation and bioflocculation, and attempts to outline new dimensions of research for a practicable application of this technology in mineral industry.

Jan 1, 2014

By R Crosbie, K C. Runge, E Tabosa, J K. McMaster

Flotation feed density is an operating parameter which is currently not well understood. Most models of flotation only incorporate its effect on particle residence time, but it is likely to have additional effects within both the pulp and froth phases of the machine. Test work was performed in MetsoÆs 3 m3 RCS test rig to investigate the effect of flotation feed density on metallurgical performance. The rig was operated at Rio TintoÆs Northparkes operation using flotation feed and scavenger tail. For each feed, tests were performed at different feed densities, and measurements were performed to estimate the pulp and froth recovery as well as other key flotation parameters (eg bubble size, turbulence and froth stability). Results of this test work have shown that the pulp density does affect key parameters of the flotation process in both the pulp and froth phases. Pulp phase kinetic rates decrease at high density and this is attributed to a reduction in the turbulence within the cell. Froth phase recovery, however, improves presumably because of a stabilising effect due to a greater loading of solids on bubble surfaces.There is a need for more research to be done to better understand these interactions so that metallurgists can determine the optimum flotation feed density to use at a particular operation, or within a section of a flotation circuit.CITATION:Runge, K C, Tabosa, E, Crosbie, R and McMaster, J K, 2012. Effect of flotation feed density on the operation of a flotation cell, in Proceedings 11th AusIMM Mill OperatorsÆ Conference, pp 171-178 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Oct 29, 2012

By M. A. Mahboob, T. Celik, B. Genc

Mineral Resources estimation plays a crucial role in the profitability of the future of mining operations. The conventional geostatistical methods used for grade estimation require expertise, understanding and knowledge of the spatial statistics, resource modelling, geology, mining engineering as well as clean validated data to build accurate block models. However, the geostatistical models are sensitive to changes in data and would have to be rebuilt on newly acquired data with different characteristics, which has proved to be a time-consuming process. Machine learning methods have in recent years been proposed as an alternative to the geostatistical methods to alleviate the problems these might suffer from in Mineral Resource estimation. In this paper, a systematic literature review of machine learning methods used in Mineral Resource estimation is presented. This has been conducted on such studies published during the period 1990 to 2019. The types, performances, and capabilities, of several machine learning methods have been evaluated and compared against each other, and against the conventional geostatistical methods. The results, based on 31 research studies, show that the machine learningbased methods have outperformed the conventional grade estimation modelling methods. The review also shows there is active research on applying machine learning to grade estimation from exploration through to exploitation. Further improvements can be expected if advanced machine learning techniques are to be used.

Nov 30, 2022