Search Documents

By K. Baba

Liberation of valuable minerals from a gangue matrix during size reduction of multiphase mineral particles is an important objective in mineral separation processes. A methodology for the prediction of volumetric grade distribution from one dimensional, linear intercept image analysis is being developed using a transformation function to describe the relations hi p between the linear and volumetric grade distributions. As part of this research effort it has been found that a transformation function, derived from Monte Carlo simulations of irregularly shaped particles, mainly depends on the volumetric grade and the dispersion density, i.e., the number of nuclei (or grains) per particle. A possible mathematical form for the transformation function is described in detail.

Jan 1, 1985

By R C. A Minnitt

The two methods for calibrating the parameters K and (alpha) for use in Gy’s equation for the fundamental sampling error (FSE), the duplicate sampling analysis (DSA) and the segregation fractionalised analysis (SFA) are described in detail. A case study using identical broken reef material from a Witwatersrand-type orebody is calibrated using the DSA and SFA methods and the results are compared. Classically, the form of Gy’s equation for the FSE raises the nominal size of fragments given by dN to a power of 3. Gy’s formula is a theoretical formula applicable in the case when fragments are randomly collected one by one, which is obviously never the case. This is because the fragment shape factor (f) and the granulometric factor (g) were introduced to take care of this anomaly. More recent formulae, such as those suggested by François-Bongarçon and Lyman, are empirically based formula. A later modification of Gy’s formula raises dN to the power ? (alpha), the latter term being calibrated with the coefficient K in the DSA and SFA methods. The preferred value of ? (alpha) for low-grade gold ores used by sampling practitioners in the mining industry is 1.5. A review of calibration experiments for low-grade gold ores using the DSA and SFA methods has produced values of K that vary between 70 and 170 and values of alpha in the range 0.97 to 1.30. The average value for alpha is found to be 1, rather than 3 as originally proposed in the classic form of Gy’s equation or the industry preference for 1.5. It is suggested that for low-grade gold bearing ores, the equation for the FSE should raise dN to a power of 1. Such an equation for the variance of the combined FSE plus the grouping and segregation error, now referred to as the short-term component of the quality fluctuation error (QFE1), greatly simplifies the characterisation of gold ores, which now only requires the calibration of K for a given mass and established fragment size. The implications for the heterogeneity test from the generalised equation are that provided the fragments have been screened to within a narrow size range, any particular size will return a value for K that is acceptable for use in the sampling nomogram.CITATION:Minnitt, R C A, 2017. A generalised form of Gy’s equation for gold ores – empirical evidence, in Proceedings Eighth World Conference on Sampling and Blending , pp 331–350 (The Australasian Institute of Mining and Metallurgy: Melbourne).

May 9, 2017

By J. C. Johnson

The problem of estimating pillar capacity in coal mines has motivated development of a variety of empirical equations. Most of these equations have arisen from experience and a very limited set of measurements that have, necessarily, been focused on particular regions or coal fields. Their applicability is often limited to a specific range of pillar sizes. This is of particular concern in the design of large gateroad pillars and small barrier pillars between panels. The result has been a variety of equations and equation coefficients that can calculate load estimates to size pillars. Inclusion of these equations in more sophisticated loading models, such as boundary element methods (e.g. LaModel and MULSIM) has been hampered by absence of a function that describes specific variation of coal strength within the pillar. A singular exception, to our knowledge, is a function corresponding to the Bieniawski pillar strength equation. Because of this, it is the only coal strength criteria inherently available in most boundary element models. This paper presents a general method for deriving local coal strength, as a function of distance from the pillar rib, for an arbitrary pillar strength equation. Integration of this function provides the strength of pillar regions, or cells, needed for boundary element models. The method is demonstrated for the Bieniawski, Maleki and Holland-Gaddy equations. The Maleki and Holland-Gaddy equations contrast strongly with Bieniawski as strengths are extrapolated to large pillars and barriers. This development provides a tool for incorporating any of the proposed empirical equations into a boundary element model, thereby allowing the analyst to choose the most appropriate equation for a particular region, coal seam and/or pillar size.

Feb 23, 2014

By Ralph W. Barbaro

In the mineral industry, the objective of a mining venture is to extract, process, and market a mineral commodity in such a manner that the profit of the venture is maximized. In a mining venture, a company normally has control over mine production scheduling, the selection and location of processing facilities (mineral beneficiation plants, smelting plants, refineries, or blending facilities), processed ore blending, and the selection of markets. The flexibility of the above factors provides many alternatives to satisfy the needs of each market with regard to quantity and quality of ore. The purpose of this paper is to present a mathematical model for determining the alternative that yields the maximum profit. Production planning is generally concerned with the best utilization of fixed resources, more specifically, with the determination of production and inventory levels to meet fluctuating demand requirements. A mine can be any source of ore such as an entire mine, a production unit within a mine, or an independent supplier. The objective of production scheduling is to determine the times at which each mine will be used and to allocate a required level of production. Production planning and scheduling problems are intimately associated with the location of new and existing mines and processing facilities. Although the location of the mines is usually fixed, there can be considerable flexibility in locating new processing facilities, in the selection of the processing plant, and in blending the ore before and after processing.

Jan 1, 1983

By Markus A. Reuter, Jacques J. Eksteen

The modelling of bath type furnaces tends to be either fundamental in nature or total empirical "black box models". While models of the first kind are robust, they tend to lack accuracy, in cases where they are to be used for process control, and may take an overly long in development time. On the other hand, black box models tend to lack robustness of prediction for all circumstances and tend to be limited to interpolate based on historical data. While this may very often provide short-term accuracy, the models may have problems with stability in the long term. Moreover, the model structure may not reflect any of the fundamental properties of the system and remains opaque to fundamental analysis. The aim of the model development is in obtaining a model that have robustness, high accuracy, a short development time and which is adaptive to a changing environment. The models are specifically developed for on-line decision support or control of furnaces. The modelling approach is evaluated for a chromite bath smelting furnace. The role of systems analysis, mass balance smoothing, thermodynamic analysis are integrated into one generic algorithm for bath type furnaces. Therefore, this paper argues that to understand complex metallurgical furnaces a hybrid approach is the only method to fully grasp the intricacies of metallurgy, to improve operations, economics as well as environmental performance.

Jan 1, 2003

"This generic study has been developed in response to the frequently asked question, ""That is the best treatment route for refractory gold ore?"". The study has developed operating and capital costs for many potential treatment routes and they are presented here in a generic form. All costs are in relation to the value of 1.00 given to the base case, a 1000 mtpd bio-oxidation plant, operating at 25% solids and 72 hours residence time.There is unfortunately no simple answer to the above quest ions. The treatment of refractory ores is a complex subject, with many variables, such as mineralogy, chemistry, environment and tonnage. However this study does offer information and relative numbers that we believe will be of assistance to gold metallurgists in obtaining a ""first shot"" at estimating costs.Equally, this technology is a rapidly evolving one. As more plants are commissioned, and test programs completed, better data is constantly being generated. For a generic study of this type to remain useful it will need regular updating and adjustment."

Jan 1, 1989

By C. Musingwini, A. S. Nhleko, P. M. Muke

Open-pit mine planning includes the determination of a long-term open-pit mine production scheduling (OMPS) plan. The OMPS plan is generally a three-dimensional (3D) block extraction sequence that seeks to maximise the net present value (NPV) of future cash flows, while complying with technical and operational constraints. The OMPS problem is computationally complex because it involves the analysis of large datasets, while simultaneously incorporating both multiple hard and soft constraints. The OMPS problem has been solved using deterministic or exact approaches, heuristic, and metaheuristic models. When compared to heuristic and metaheuristic models, deterministic approaches tend to take much longer to solve the OMPS problem when the number of constraints increases. Therefore, this paper presents a metaheuristic model which utilises a genetic algorithm (GA) framework to solve the OMPS problem. The GA-based model was applied to the Newman1 and Zuck Small mining instances, which are publicly available from the online MineLib database. The NPVs for the OMPS plan generated using the GA-based model for the two case studies were compared to the best-known published MineLib solutions obtained using the TopoSort heuristic model at the time of this study. For the Newman1 instance the equivalent annual NPVs were approximately US$6.6 million for the GA-based solution and US$5.1 million for the TopoSort solution. For the Zuck Small mining instance, the equivalent annual NPVs were approximately US$99.8 million for the GA-based solution and US$102.5 mil for the TopoSort solution. These results show that the GA-based model produced comparable solutions to some of the best-known solutions, thus, indicating strong potential application of GA-based models to solve OMPS problems.

Sep 1, 2021

By Biswajit Samanta, Pranjal Pathak

This article presents a customized genetic algorithms (GA) with new crossover and mutation operators to solve the short-term production scheduling problem of multi-mine mineral value chains (MVC). The preceding problem consists of determining the extraction sequence and destination allocation of blocks from all the mines collaboratively while closely meeting the quality and quantity requirements of the processing units subject to relevant technical and operational constraints. The short-term production scheduling is carried out at shorter scales wherein the operations are modeled in a great detail with large number of constraints. This makes the industry-scale instances of the problem computationally intractable for standard mixed-integer programming (MIP) solvers. Thus, a GA-based heuristic approach is developed to obtain nearoptimal solutions to the large-scale instances of the problem in a reasonable amount of computational time. Computational experiments show that the developed GA-based method is a promising way to handle industry-scale instances of the problem. Moreover, the sensitivity analysis on various parameter combinations of crossover and mutation operators indicates that the customized global mutation operator, when used in combination with the customized crossover operators, took on average 12.5% less time than the customized local mutation operator to converge to a solution.

May 21, 2022

By Peter A. Hacquebard

Recent drilling programs have shown that the in-situ coal resources of Nova Scotia are very substantial. The grand totals of the measured, indicated and inferred categories amount to 232 m, 646m and 993 m tons, respectively. This vast quantity of coal occurs essentially in five main fields: Sydney, Mabou, Port Hood, Pictou and Springhill. The largest resources (84% of the total) occur in the submarine area of the Sydney coalfield. With a 65% recovery rate, there are at present 150m tons of proven, recoverable coal in Nova Scotia. At a predicted maximum annual production of 5 m tons, this amount equals a lifespan of 30 years. However, additional drilling, particularly in the Sydney offshore area, holds promise of increasing the recoverable reserve tonnage to at least 350 m tons. This figure may be compared with the 484 m tons production obtained during 191 years of mining (from 1785-1976). The geological appraisal of the remaining coal resources in the five fields mentioned is presented in three-dimensional diagrams, per individual seam and corresponding reserve area. In addition, the submarine extensions of the Sydney and Mabou coalfields are discussed in light of results obtained from off shore oil exploration wells. The terminology and standards of coal resource calculations are pointed out, and the phenomenon of seam splitting and its prediction from geological data is illustrated. The availability in Nova Scotia of either thermal or metallurgical coal is considered in relation to rank and sulphur content. This paper was presented in June 1977 at the 90th Annual Meeting of the Mining Society of Nova Scotia. The coal resource estimates, however, have been updated to include the results obtained with the offshore drilling in the Sydney Coalfield to November 1977.

Jan 1, 1979

By P J. Mienie

The Palaeo-Proterozoic Transvaal Supergroup in the Northern Cape Province of South Africa is host to the largest known resources of high-grade hematite ore on the Southern African continent. These ores are currently being exploited at Sishen and Beeshoek. Hematite occurances on the Sishen South exploration project (some 65 km south of Sishen mine), are at present the focus of an intensive investigation. At both the Sishen and Sishen South deposits, mineralisation is found within lithologies belonging to the Transvaal and Oliphantshoek Supergroups. The Transvaal Supergroup comprises a basal, carbonate platform sequence (the Campbell Rand Subgroup) overlain by thick units of banded iron-formation belonging to the Asbestos Hills Subgroup. The Asbestos Hills Subgroup is unconformably overlain by sedimentary rocks of the Gamagara Subgroup. The latter Subgroup has been correlated with the lowermost unit of the Oliphantshoek Supergroup. The Sishen and Sishen South deposits are located close to the western margin of the Kaapvaal craton and have been subjected to intensive, structural deformation. Both localities have undergone folding, faulting and thrusting. Early regional tectonism (2400 - 1700 Ma) played a critical role in the genesis and preservation of the ores from younger erosive events. At both deposits, four similar iron ore occurrences have been recognised. The ores have developed in specific depositional environments and hence display unique physical, chemical and metallurgical properties:The bulk of the high-grade hematite comprises laminated and massive ores, which are found in the upper parts of the Asbestos Hills Subgroup. The laminated and massive ores are best developed in basinal and pseudo-graben structures. Collapse breccia ores of medium- and lower-grade are preserved within deep palaeosinkholes. The sinkholes have developed within the dolomites of the Campbellrand Subgroup. Lower-grade conglomeratic and gritty ores constitute the remainder of the deposits. These ores are found in the overlying clastic succession of conglomerates, shales and quartzites belonging to the Gamagara Subgroup. Although, there are many stratigraphic and lithological similarities between the Sishen and Sishen South deposits, there are also important differences, which will impact on the profitable exploitation of the ores. The most striking contrast between the deposits is their size and geometry. Sishen mine is a single pit operation with an opencast mineable reserve of 895 Mt. The Sishen orebody is thicker, more continous and significantly larger than the Sishen South deposit. The Sishen South deposit has a potential opencast reserve of 228 Mt. It comprises a number of small and narrow, isolated orebodies, spread over an area of ¦64 km2 . There is a significant difference in grade between the deposits. On average the orebodies of the Sishen South deposit have a higher iron content (+0.25 per cent Fe), less potassium (- 0.090 per cent K2O) and less phosphorus (-0.019 per cent P). The Sishen South deposit has less interbedded shale in the laminated and massive ores. The footwall of these high-grade ores is a light-density, intrusive sill, compared to banded iron-formation at Sishen. With a proposed heavy-medium separation plant at Sishen South, the impact of dilution on product grades is expected to be less than that currently experienced at Sishen.

Jan 1, 2002

By Carl Lausen

Within recent years several mining companies have sent expeditions co the east end of Great Slave lake co determine the mineral possibilities of that area. Little was known of the geography of the region, and even less of the geology; and, not until Blancher's survey was made and the results published, was an accurate map of the lake available. Early explorers followed the Inconnu channel and the north shore of McLeod bay to Port Reliance, and, just prior to Blancher's survey, Hornby showed that the southern passage was feasible. Apparently, but few geologists have ? travelled through the country, for geological information is meager. In 1900, Robert Bell, of the Geological Survey, made a trip to Fort Reliance by way of the north shore of McLeod bay, and he gives a brief description of the geology. The present paper is the result of an expedition into this little-known region by Mr. Geo. M. Douglas and the writer in the summer of 1928. The south shore of Great Slave lake was followed to the east end of Tochatwi bay, and an extensive exploration was made of Kahochella and Pethei peninsulas. The prime object of the expedition was co determine the presence or absence of the copper-bearing Keweenawan basalts.

Jan 1, 1929

Many aspects of best practice resource evaluation do not create value, but poor practice can certainly destroy value. Generally companies prefer to present how their good practice adds value and for this reason few publications exist on examples of how poor practice can destroy value. However in big and small companies there are many previous and current examples of poor practice destroying value. The incremental expenditure required to move from poor practice towards best practice can be looked upon as insurance against destroying shareholder value. Examples of how poor practice in geology has destroyed value are presented as well as the improvement model used by BHP Billiton to add value and insure against destroying value.

Jan 1, 2002

By T Muzondo, C Prins, D De Vaux, R Schouwstra

The Mogalakwena Platinum Mine in the northern limb of the Bushveld Complex is Anglo American PlatinumÆs only open cast operation exploiting the Platreef. The Platreef, which extends along the basal contact of the Bushveld Complex, is geologically and mineralogically a much more complex orebody than the Merensky and UG-2 reefs. Changes in the footwall and the genesis of the Platreef translate into a number of rock types being present. These rock types vary in hardness and recovery potential and present challenges to metallurgical staff trying to optimise the concentrator. A true understanding of the implications of treating these rock types is required if the maximum value from this orebody is to be obtained.In order to more accurately forecast production and incorporate ore variation in the mine plan and comminution and flotation circuits, a geometallurgical approach to managing the operation has been used and is being further developed.CITATION:Schouwstra, R, De Vaux, D, Muzondo, T and Prins, C, 2013. A geometallurgical approach at Anglo American PlatinumÆs Mogalakwena Operation, in Proceedings The Second AusIMM International Geometallurgy Conference (GeoMet) 2013 , pp 85-92 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 29, 2013

By B J. Beven, A Vatandoost, J Young, S Campbell-Hardwick

The intrinsic value of iron ore depends on its physical and chemical characteristics. For example, silica and alumina are unwanted elements, downgrading iron grade and adversely affecting high temperature processing, attracting penalties in smelter contracts. For this reason, an informed knowledge of processing characteristics and product quality of the ore is valuable and cost effective during mine planning and processing.Ore variability is a significant factor that affects metallurgical behaviour and efficiency of ore processing equipment. Ore of sufficient chemical and physical quality needs to be modelled, mined, processed and usually blended. Feed and product assays are commonly used to assess potential ore upgradability. However, assays alone cannot be all informative of actual processing response given the variability of the physical properties of the ore.This study investigates the potential benefit of implementing a geometallurgical approach during the commissioning stage of the Kings deposit, part of the Solomon Iron Ore project in Western Australia. Kings drill hole database information, ie geological, textural and geochemical data, together with the results of metallurgical test work were used for geometallurgical analysis and modelling. Three main ore types were identified at Kings, namely channel iron deposit (CID), bedded iron deposits (BID) and Detrital Iron Deposit (DID). Additional classifications were utilised to better characterise these units in terms of their physical properties.Bulk sampling for each ore type was carried out across the deposit for metallurgical testing. Multivariate statistical models were developed for prediction of product quality for each geometallurgical unit. A new approach for modelling yield based on ore upgrade factors is introduced. These models were then applied to the Kings regularised mining model to assess product quality across the mining blocks and to plan product stockpiling and blending strategies.CITATION:Vatandoost, A, Beven, B J, Campbell-Hardwick, S and Young, J, 2013. A geometallurgical approach for iron ore product evaluation, in Proceedings The Second AusIMM International Geometallurgy Conference (GeoMet) 2013 , pp 259-266 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 29, 2013

By R Schmitt, Y Ghorbani, I McElroy, N J. Bolin, M Parian

Aitik is a large copper porphyry type deposit located in northern Sweden, currently exploited at an annual rate of approximately 45 Mt. The ore’s exceptionally low head grade of 0.25 per cent Cu and varying degrees of hardness across the entire deposit pose challenges to the two fully autogenous grinding lines, each of which comprises a primary autogenous mill in series with a pebble mill. The variability in ore grindability frequently leads to fluctuations in mill throughput. Within the framework of a geometallurgical approach, the present study investigated the relationships between ore grindability and modal mineralogy. For this purpose, drill core samples from different lithologies were subjected to Boliden AB’s in-house grindability tests. This fully autogenous laboratory-scale test generates a grindability index mainly related to abrasion breakage, which is a significant breakage mechanism within autogenous mills. The test results suggested divergent degrees of grindability within and across the selected rock types. A combination of scanning electron microscopy, X-ray powder diffraction, and X-ray fluorescence analyses was performed for the grinding products and bulk mineral samples. The resulting mineralogical and elemental properties were subsequently correlated to the parameters from the grindability tests. It was shown that the main mineral phases, such as plagioclase, quartz, and micas, correlate well with the grindability indices. Similar correlations were found regarding the sample’s chemical composition, attributable to the main mineral phases. A further inverse correlation was established between the sample’s calculated average Mohs hardness and the grindability indices. Moreover, mineral liberation information provided by scanning electron microscopy was associated with the parameters mentioned earlier. The identified relationships between grindability, modal mineralogy, and element grades may help Boliden develop a predictive throughput model for Aitik based on the mine’s block model.

Aug 24, 2022

By D L. Reid, G Nwaila, J Petersen, J-P Franzidis, Y Ghorbani, M Becker, F C. de Beer, L C. Bam

Traditional methods of gold extraction have involved comminution and processing via cyanide leaching, all of which are associated with energy, cost and environmental implications. Subsequently, numerous opportunities for improvement and optimisation strategies in gold mining and processing have been explored, including heap leaching, which has significantly lowered comminution costs. This paper explores the effect of the more energy efficient high-pressure grinding rolls (HPGR) on the comminution behaviour (especially on the crack density) and cyanide leaching performance of laboratory reactors study on run-of-mine gold ore comprising the Ventersdorp Contact Reef (VCR) and the Elsburg Reefs (ER), sourced from Carletonville in the Witwatersrand Basin in South Africa. The ore was comminuted using HPGR at three different pressure settings (120, 90 and 60 bar). Large (-16/+11.2 mm) and small (-5.6/+4 mm) particles of the HPGR product where then mineralogically characterised using optical microscopy, quantitative evaluation of minerals by scanning electron microscopy (QEMSCAN) and X-ray computer tomography (CT) before cyanidation. The results of the leach experiments indicated a correlation between gold extraction, particle size and crack density. Mineralogical characterisation of the ore indicates that the Fe-sulfides present in the VCR are likely to be more reactive than those in the ER, thus having a more favourable influence on gold cyanidation. Overall, the results show promise for the heap leaching of Witwatersrand gold ores prepared using HPGR and demonstrate the value of a multi-disciplinary geometallurgical approach.CITATION:Nwaila, G, Becker, M, Ghorbani, Y, Petersen, J, Reid, D L, Bam, L C, de Beer, F C and Franzidis, J-P, 2013. Morro do Ouro geological model with a metallurgical view, in Proceedings The Second AusIMM International Geometallurgy Conference (GeoMet) 2013 , pp 75-84 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 29, 2013

Designing rigging systems for complex castings has been more of an art than a science. Foundrymen have used thumb rules based on past experience to develop working designs. Of late, attempts are being made to formalize the procedure through the use of rule based expert systems, process modeling and other tools. Designing risers is one of the challenges of the rigging system design. Once the riser location have been determined, an optimization technique coupled with a casting simulation code can be used to determine the optimum riser dimensions. The main drawback of using a finite element or a finite difference based process simulation code along with the optimization routine is that the computation time involved is too large. Such an approach, thus, is unrealistic for practical situations where different options have to be explored and conclusions drawn in a short period of time. This difficulty has been avoided in the present work. The optimization scheme has been coupled with a modulus based solution routine to optimize the riser dimensions. The modulus calculations provide a good and quick estimate of the freezing time at different points in the casting. A functional, obtained from these freezing time values, is minimized by altering the dimensions of the riser until an optimum shape is obtained. Some examples have been included to illustrate the utility of this technique.

Jan 1, 1994

The word "optimization", refers to the search of an economic optimum, or, more precisely, to the choice of technical parameters which will define the more profitable mining project, knowing the values of the economic parameters. In fact, there is no direct way to perform this optimization, because the extremely large number of technical factors involved, from the type of working and its selectivity to the characteristics of the mill and its flexibility, makes it impossible to make a comprehensive survey of all possible projects. Beside, it would be meaningless to engage in such a task as the deposits themselves are not really known, but only estimated, with sometimes a high level of imprecision. Not to mention the numerical instability which would affect such a computation! In addition, the "best" project is not necessarily the one which strictly maximizes a given profit formula. There also exist human factors which have to be complied with, when choosing the cutoff grades for instance. Finally, there are other non-numerical factors which interfere with the choice between several projects: the shape of the bottom of a pit of the layout of the access road for instance.

Jan 1, 1986

By J F. C L Costa, J C. Koppe

It is recognised by the major international codes that resource classification involves the interaction of numerous qualitative and quantitative criteria such as data quality, geological and grade continuity, etc. However, the difficulty in quantifying the degree of uncertainty associated with mineral resources estimation has led to the creation of a large suite of methodologies, terms and definitions, with almost every mining company having its own set of standards. It is also intuitively known that several traditionally used methods to evaluate resources do not take into account the spatial continuity of the grades, being unable to provide a reliable measure of the error associated with their estimates. Since they do not provide an error assessment, these methods are inappropriate to assess local or global uncertainty associated with an estimate. Faced with these problems, this work aims at developing a methodology for either quantitative or qualitative analysis of mineral resources estimation, through incorporation of uncertainty and the correct definition of the associated risk or error. The proposed methodology includes the development of a software code incorporating all the main classification techniques used by the mineral industry. A comprehensive study about each technique was conducted allowing a comparison of the parameters affecting mineral inventory assessment. A case study was conducted with a typical iron ore deposit data set from the Conceitpo mine of the Itabira Complex in the Iron Quadrangle, Brazil. The results showed the specific limitations of each classification system and the influence in selecting their key parameters, always based on subjective choices.

Jan 1, 2009

By G. Pan

Although geostatistical techniques have been widely accepted as the standard methods for reserve calculations in the mining industry, the production data of many mines frequently differ from geostatistical estimations based on exploration and development drilling. The discrepancies in recovered metal amounts, commonly in the range between 10% and 30%, are related to different factors. Major causes vary from mine to mine. A common cause, however, is the ignorance of important geological constraints in reserve modeling. Three major factors appear to be related to such an unfortunate fact: 1) modelers are over-confident on geostatistical techniques, 2) it is difficult to establish geological models, and 3) mine schedules are too tight to allow detailed geological modeling work. Defining mineralization envelopes has been a major effort in geological modeling. Past experiences have showed that geological modeling is much mo~ demanding than the choice of sophisticated kriging estimators. Since building envelopes for complex ore bodies can be extremely tedious and time-consuming, computerized approaches are badly needed for mine modelers. A comprehensive procedure reported here is referred to as probability-assigned constrained kriging (PACK), which includes a built-in method for defining 3-d mineralization envelopes. The modeling technique is fast and accurate for complex precious metal deposits. The technique is demonstrated on a gold deposit. The estimation results are compared with blast-hole based production panels, showing that the estimated tonnage is 2.2% larger than the actual production and the average grade is 2.6% lower.

Jan 1, 1994