Search Documents

By P. S. Bagga, P. T. Luckie

Liberation (the process of destroying the interlock between unwanted materials, such as mineral matter and pyrite, and coal) is one of the most important precursors to the benefication of raw coal in order to produce high grade marketable coals at economical recovery levels. Liberation is achieved usually by comminuting the raw coal. However, because of the significant interaction between the production of fine particles to achieve liberation and the cost of separating fine particles, the coal industry needs a method projecting the change in washability with size reduction. Such a projection can be combined with size reduction and separation simulators in order to examine completely beneficiation circuitry to produce coals of stringent specification. This paper presents the development of an appropriate liberation model for coal and its solution by Monte Carlo simulation to predict the expected washability of the raw coal as a function of comminution. It includes model validation with a known binary mixture as well as with a known teritary mixture.

Jan 1, 1983

By A. Muwila

Hercules? is a low nickel austenitic stainless steel developed at Mintek that has a typical composition of 9 wt% Mn, 0.05 wt% C, 2wt% Ni, 0.25 wt% N and 16.5 wt% Cr. A more corrosion resistant version of Hercules? (Hercules? B) with 0.5 wt% Mo has also been developed. Potentiodynamic tests performed on Hercules? B in a 5wt% sulphuric acid solution showed it to have improved corrosion resistance comparable to that of AISI 304. The effect of a varying Mn content on the corrosion resistance of Hercules? B has also been investigated and the results discussed in this paper. From the corrosion test results, it is clear that decreasing the Mn content of Hercules? B increased the corrosion rate, while increasing it decreased the corrosion rate. Keywords: austenitic stainless steel, % ferrite, molybdenum, manganese, nitrogen, electrochemical corrosion tests, potentio-dynamic scans, microstructure, chemical analysis.

Jan 1, 2007

By M. J. Papo, A. Muwila

Hercules? is a low nickel austenitic stainless steel developed at Mintek that has a typical composition of 9 wt% Mn, 0.05 wt% C, 2 wt% Ni, 0.25 wt% N and 16.5 wt% Cr. A more corrosion resistant version of Hercules? (Hercules? B) with 0.5 wt% Mo has also been developed. Potentiodynamic tests performed on Hercules? B in a 5 wt% sulphuric acid solution showed it to have improved corrosion resistance comparable to that of AISI 304. The effect of a varying Mn content on the corrosion resistance of Hercules? B has also been investigated and the results discussed in this paper. From the corrosion test results, it is clear that decreasing the Mn content of Hercules? B increased the corrosion rate while increasing it decreased the corrosion rate. KEYWORDS Austenitic stainless steel, %ferrite, molybdenum, manganese, nitrogen, electrochemical corrosion tests, potentiodynamic scans, microstructure, chemical analysis.

Jan 1, 2006

By N Weerasekara, M Bird, M Yahyaei, M M. Hilden, M S. Powell, J Franke, P Toor

The combination of new tools with a more rigorous approach to understanding the interaction of mill liners with overall mill performance and economics, forms the basis of an integrated approach to liner management. Based on accurate liner wear measurements, some (limited) physical mill inspections and surveys, accurate calibration of the mill load and analysis of process data, a whole new approach to liner selection, relining intervals and mill control is proposed. Collection and analysis of this information in operating mills illustrates the mill performance improvement which can be attained through this careful measurement and modelling approach.A more holistic approach to liner management is proposed, including a broadening of the current work to cover the potential benefits of improved relining technologies and predictive liner wear modelling. CITATION:Powell, M S, Hilden, M M, Weerasekara, N, Yahyaei, M, Toor, P, Franke, J and Bird, M, 2012. A more holistic view of mill liner management, in Proceedings 11th AusIMM Mill OperatorsÆ Conference, pp 95-104 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Oct 29, 2012

By P. Wells, P. Quinn

"A moveable flotation miniplant was designed and constructed at the research laboratory of Inco Technical Services Ltd. to address the need for flowsheet development with a limited amount of ore (such as drill cores). It consists of one rod mill, two regrind ball mills, one vibratory screen, 24 flotation cells of various sizes and two Venturi aerated flotation columns. The setup has continuous monitoring capability for pH, dissolved oxygen, oxidation-reduction potential, conductivity and temperature in conditioning tanks and flotation cells. Various options regarding flotation reagents, circuit configurations, and fineness of primary grind and regrind can be readily evaluated. The throughput of the miniplant is 20 – 30 kg/hour ore.Unit operations such as flash flotation, magnetic separation and gravity concentration can also be evaluated as an integral part of an entire flowsheet development.This communication describes the basic design features of the miniplant and some examples of applications are given.INTRODUCTIONInco operates numerous mines in Sudbury and Thompson and will shortly bring the Voisey’s Bay mine into production. It has long been recognized that the ores from different mines or from different depths of the same mine can give different responses in flotation. There is a growing need to properly evaluate the ores before feeding the mill to maximize the return of the mine/mill operation. Inco Technical Services Limited at Sheridan Park, Mississauga, Ontario has developed a batch laboratory procedure to address this issue. The results are considered in the overall economical assessment of new mine operations.For ores that cannot be processed by the existing mill operations or that will be processed by a new mill, it is necessary to develop a flowsheet that can maximize the economical return of the mine/mill operation. While recognizing the importance and contribution of the laboratory batch flotation procedure, it does not generate reliable data for flowsheet simulation and design. A full-scale pilot plant campaign is normally required to obtain the design data but often this is not possible due to a limited amount of ore obtainable from the drill core. A small-scale pilot plant (mini pilot plant or miniplant) capable of processing 20-30 kg/h ore can provide this data prior to a full-scale pilot plant campaign. A miniplant can also be used to optimize the circuit configurations, reagent additions and recycle options of process water before pilot plant work and can be used to test a different flowsheet configuration for an existing mill before re-piping a plant circuit."

Jan 1, 2003

By R. Demont, A. Pranesh, S. Rakheja

Ethical concerns of in-vivo procedures and poor repeatability of non-invasive techniques have been major limitations in estimating vibration-induced spine loads through experiments. The biodynamic models of seated human body exposed to whole-body vibration (WBV) have evolved for defining the frequency-weightings, enhancement of human responses to WBV, and developing anthropodynamic manikins for seating assessment activities. The widely reported mechanical-equivalent models, solely based on through- or to-the-body biodynamic response functions, do not seem to resemble the biomechanical structure and do not yield information on the dynamic loading and deflections of segments of concern, namely the spine. On the other hand, biomechanical models with representative anatomical structure and anthropometry are being attempted to simulate segmental movements and the coupling effects, using Finite elements (FE) or multi-body dynamics (MBD) formalisms, which could provide important insights into the inter-vertebral forces [1]. While the FE models pose considerable complexities primarily related to characteristics of the bio-material properties, the MBD technique with discrete rigid bodies offers the flexibility to create multi-segment models with relative ease and lower computational cost. In this study, a preliminary multibody dynamic model of a seated human body exposed to WBV along the vertical direction is formulated using MSC/ADAMS software. The model validity is demonstrated by comparing selected responses with the available measured data. Methods The seated human is represented by nine rigid body segments, including: head, neck, thoracic and lumbar torso, pelvis, hands and thighs, as shown in Fig. 1. The rigid bodies are coupled through different rotational and translational joints, some of which are force elements to allow vertical translations and sagittal–plane rotations of the segments. The measurements of transmission of vertical vibration through-the-body generally require subjects to voluntarily maintain a vertical head position to reduce head-accelerometer orientation errors. The head–neck–shoulder joint is thus considered to be rigid. The shoulders are assumed to be rigidly attached to the thoracic segment. The torso is made up of three (upper, middle and lower) segments connected by visco-elastic revolute and translational joints to permit relative pitch and vertical motions. The forces and torques generated by the joints are derived assuming linear stiffness and damping properties, which were identified from published studies. The pelvis is connected to the rigid seat by similar elements representing the visco-elastic properties of the buttock tissues. The two thighs are rigidly connected to the pelvis, while the segment masses are chosen from the anthropometric data for the 50th percentile male subject. The initial model parameters for the joints were obtained from [2]. The model was analyzed to determine the force-motion relationship at the buttock-seat interface expressed in terms of

Jan 6, 2006

By S. A. Abdel Sabour

Conventional methods for selecting a production plan for an open pit mine is based on the assumption that ore grades throughout mine life as well as market variables are known with certainty. Another important issue that is also being ignored in the conventional procedure is the value of management flexibility to revise operating policy with time according to the arrival of new information. Based on this oversimplification of reality, an optimum mine plan that is thought to be maximizing net present value (NPV) of the project is selected. Given the multiple sources of risk affecting the estimated NPV, there is a high probability that the actual value will be different from that estimated. This could mislead decision making process regarding selecting an optimum mine plan. This paper proposes a multi-criteria ranking system for selection between alternative mine designs under both geological and market uncertainties. The system is based on integrating multiple market and geological uncertainties as well as the operating flexibility to revise the ultimate pit limits using a Monte Carlo-based real options valuation (ROV) model. To compare the methods, the article applies the proposed system, along with the conventional NPV-based method, to rank possible mine designs for a copper and a gold mine. It has been found that, considering multiple risk analysis measures while integrating uncertainty and operating flexibility results in selecting a mine design that is different from that selected by the conventional method.

Jan 1, 2009

By Jinsuo Zhang

Investment in coal resource development is affected by a variety of uncertain factors. Investors can make a flexible, strategic decision to achieve the value of risk implicated in those uncertain factors. Therefore, based on real-option theory, we identified key uncertain factors that affect a coal resource investment project, and revealed the formation mechanism of option value in coal resource developing investment project. Then, a multi-factor decision making model was established for a coal resource development investment project with random changes of coal price, interest rate, convenience yield and production cost. Finally, we tested the model and analyzed relevant parameters by applying living data. Our results indicate this model reflects the value of risk implicated in uncertain factors for coal resource development investment, and is able to determine a critical point of decision making change when coal price and production cost varym respectively. By means of parameters analysis, the convenience yield has a positive effect on the project value. The interest rate has a negative effect.

Aug 1, 2013

By G Hancock, J Lowry, T Coulthard, W Erskine, M Saynor

An important part of planning for the rehabilitation of uranium mine sites is the ability to assess the geomorphic stability of the final landform, over time frames ranging from decades to millenia. Depending on the environmental conditions, erosion features such as gullies may incise into containment structures, potentially leading to the exposure and transport of encapsulated radioactive material. Further, erosion may lead to increased sediment loads and the transport of other mine-related contaminants off-site and into downstream waterways. While erosion may be monitored and assessed over the short term (years to decades), the longer-term behaviour (centuries to millennia) of such landscapes is not within any meaningful human management time frame. Landscape evolution models (LEMs) can provide information on soil erosion rates at decadal or centennial scales over large spatial scales, and evaluate the sensitivity of these processes to environmental changes. However, an important issue associated with the use of models is the ability to assess the reliability and accuracy of the model. In this paper, the CAESAR-Lisflood LEM is tested for its ability to predict soil erosion from a series of 30 m × 30 m experimental plots specifically constructed for this purpose on a trial rehabilitated landform at the Ranger Uranium Mine in the Northern Territory, Australia. Model predictions of bedload and suspended sediment loads from the erosion plots were compared with field measured observations collected over four wet seasons from the trial landform. Once calibrated for the specific site hydrological conditions, the predicted sediment yield – specifically, bedload – demonstrated an excellent correspondence with the field data. However, longer-term simulations of ten years identified an exhaustion effect in sediment yield from the landform. These latter results indicate that a weathering function needs to be incorporated into CAESAR-Lisflood to ensure confidence that the CAESAR-Lisflood LEM will be able to correctly predict the long-term evolution of a rehabilitated landform once it has been?constructed.CITATION:Lowry, J,?Saynor, M,?Erskine, W, Coulthard, T and Hancock, G, 2014. A multi-year assessment of landform evolution model predictions for a trial rehabilitated landform, in Proceedings Life-of-Mine 2014 , pp 67–80 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Jul 16, 2014

By M Dietrich

The paper describes an efficient method for analysing the different attachment kinematics of excavators and other linkages by taking into account digging forces, the workspace and the stability of the machines. The different attachment kinematic systems of hydraulic excavators and a æclassical analysing methodÆ will be presented. The method was integrated as a plug-in for the multibody simulation (MBS) software MSC.ADAMS to create an easy to use solution for design engineers. A comparison of the new MBS-solution with the classical static analysis is a part of this paper and shows the advantages of the new method. This takes into account the manifold and new dynamic analysing opportunities, as well as more detailed models leading to more accurate results.

Sep 26, 2011

By D Hatton, S Wilson, F Orrante, M Taghimohammadi, F van der Walt, N Arslan, M Westcott

"Cayeli Bakir Isletmeleri is an underground copper-zinc mine in turkey. The ore is highly variable with feed grades ranging from 1–4 per cent copper and 1–10 per cent zinc. Due to this variability, ore characterisation and blending strategies are crucial for performance optimisation.This paper presents a case study for a multidisciplinary approach to process optimisation, integrating Geometallurgical and advanced control methods to achieve optimised plant performance. The results from comprehensive metallurgical tests and surveys were used to develop a mathematical model utilising SGS’s Integrated Geometallurgical Simulator® (version number 2.5.24.0). The model was employed to determine the most appropriate operational settings required to ensure that optimised grades and recoveries are achieved for each ore type.SGS’s Modular Expert Technology (MET) expert system was then implemented to target these optimum operational ranges utilising online grade and froth velocity measurements. The successful conversion and deployment of the modelled process parameters into an expert system has resulted in significant improvements in overall recovery and circuit stability.Citation:Arslan, N, Westcott, M, van der Walt, F, Hatton, D, Taghimohammadi, M, Orrante, F and Wilson, S, 2016. A Multidisciplinary Approach to the Optimisation and Control of a Copper-zinc Grinding and Flotation Circuit, in Proceedings 13th AusIMM Mill Operators’ Conference 2016, pp 115–122 (The Australasian Institute of Mining and Metallurgy: Melbourne)."

Oct 10, 2016

By Mark Lee, Isobel Yeo, Sarah Howarth, Christian Millo, Javier González Sanz, Paul Lusty, Bramley Murton, Pierre Josso

"Seafloor cobalt-rich ferromanganese crusts represent the most important yet least explored resource of ‘E-tech’ elements on the planet (Hein et al. 2003; ISA, 2008; Hein et al. 2010; Hein et al. 2013). These polymetallic deposits form a continuum from nodules rich in manganese and cobalt to crusts rich in tellurium and the heavy rare earth elements (HREE). It is this combination of traditional (base) metals and the extreme enrichment of ‘E-tech’ elements that makes seafloor ferromanganese oxide deposits particularly interesting to both science and society. Recent estimates of the global resource, based on the sparse data available, infers a dry mass of ferromanganese crusts on the seafloor of 35 x 109 tonnes (35GT). Of this, 3.7 GT is predicted in the Indian Ocean, 8 GT in the Atlantic, and 23 GT in the Pacific (Hein et al. 2013).At a global scale, the processes controlling the formation and distribution of ferromanganese (Fe-Mn) crusts are reasonably well understood (e.g. Hein et al. 2000; Verlaan et al. 2004; Cronan 2006; Halbach et al. 1989; Usui and Someya, 1997; Hein et al. 2000; Hein et al. 2003; Hein et al. 2009; Hein et al. 2010; Hein et al. 2013; Muiños et al. 2013; Marino et al., 2016). However, much of this knowledge is drawn from sparse sampling at a regional to ocean basin scale. As a result, there remains a fundamental gap in understanding of the role of local-scale factors, such as micro-topography, ocean currents and upwelling, sedimentation rates, micro-organisms, water mass composition and biological productivity, that are considered potentially crucial to controlling the growth and composition of Fe-Mn crusts. Indeed, quantitative, particularly temporal, information relating to these processes and their relative importance in deposit formation is almost completely absent. Hence, to make any significant advance in understanding these deposits require significant new research."

Jan 1, 2017

By Julián M. Ortiz

Uncertainty quantification of a spatially distributed variable at any scale can be handled through geostatistical simulation. Large computation time, storage, and post-processing of the realizations are required to obtain a final assessment of block uncertainty. Multi-Gaussian kriging is a flexible alternative to simulation. The idea is to compute the conditional distribution of uncertainty after normal score transformation of the original samples. Under the multi-Gaussian assumption, all marginal and conditional distributions are Gaussian, hence fully defined by their mean and variance. The parameters of the conditional distributions are obtained by simple kriging and can be back-transformed to the original units of the variable of interest. An estimate and any summary of uncertainty can be easily retrieved. The main disadvantage of performing multi-Gaussian kriging is that change of support is not straightforward, that is, calculating uncertainty of block grades. We propose a methodology to overcome this limitation by considering a matrix simulation to generate multiple probability fields. Each probability field is used to draw spatially correlated point values from the point-support conditional distributions, and multiple realizations of the average can be obtained. This permits the calculation of the average over the block and its uncertainty. These blocks may correspond to selective mining units or to volumes from longer production periods relevant for engineering decisions. They can even be disjoint blocks, such as when several faces are mined at the same time. POSTMG, a Fortran program to perform these calculations, is described and a case study is provided.

May 1, 2004

By A Grace, C Sennersten, A Davie, R Lyu, C A. Lindley, O Goldstein, T Thomas, L De Macedo Camargo

Block modelling is a well-established method for modelling ore reserves as a basis for strategic mine planning and mine development. Off-Earth mining, particularly of asteroids, can benefit from block models; however, the significantly different characteristics of asteroids and their methods of mining require reconsideration of the geometry and annotations of block models. Terrestrial block models consist of rectangular prisms within Euclidean coordinate systems. Their labels and parameters typically include location within a mine-specific coordinate system, x, y and z dimensions, specific gravity, hardness, grades of metals of interest and so on.These data provide an input to mine design and strategic mine planning. Mining asteroids may require different coordinate systems (eg spherical, elliptical or cylindrical, rather that Euclidean) and attributes that vary from Earth-based block models. These include: variable gravitational vectors throughout the block modelrotational moment or accelerationrotation rates. Materials within asteroids may show different isotopic compositions. Asteroid mining has the unique characteristic of a mine being identical with an asteroid. The position of the asteroid/mine in space and its full orbital elements provide critical inputs to mission design and costing. Hence, the block model concept needs to be extended with this metadata. The orbital position and behaviour of the Earth must be included for active mine planning, and increasing automation of mine planning and operations needs to consider terrestrial weather and telecommunications infrastructure. Financial planning requires analysis of the net present value of extraterrestrial mining compared with terrestrial alternatives, so a fully automated mining enterprise should integrate terrestrial and extraterrestrial block models at different scales and resolutions. Blocks can be indexed via octrees, allowing for hierarchical scales that support analytics for deriving new knowledge and correlating features across different scales.CITATION:Lindley, C A, Sennersten, C, Davie, A, Goldstein, O, Lyu, R, Grace, A, Evans, B, Pentecost, D, Thomas, T, De Macedo Camargo, L, Craig, J, Salomao, G N, Hellicar, A, Biggins, D, Timms, G, Coombe, M, Shahriar, M S, Smith, G and Rahman, A, 2015. A multilayer 3D index tool for recursive block models supporting terrestrial and extraterrestrial mine planning, in Proceedings Third International Future Mining Conference, pp 289–296 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Nov 4, 2015

By Ruilin Yang

This paper presents a fragmentation model for large production blasts. The model uses peak particle velocity (PPV) induced from the blast as a controlling parameter to predict the fragmentation size distribution. The PPV can be related to shock pressure or strain under a given rock condition. The fragmentation size is calculated at three-dimensional grid points within the blast volume. The grid size defines the calculation resolution of the model. The fragment size distribution is assumed to be in the same form as the crack distribution within a cube (of the size equals to the grid size) surrounding a grid point. The crack distribution in rock from impact is based on the finding from previous researchers (Seaman et al., 1976). The fines and oversized blocks are calculated explicitly. The summation of the size distribution at all calculation grid points over a specified volume of the blast yields an overall size distribution for the volume.

Jan 1, 2015

By R Yang

This paper presents a fragmentation model for large production blasts in an open pit or underground environment. The paper shows that the particle velocity is directly related to pressure, strain, and strain rate and so it can be considered as driving force and mechanism of rock fragmentation. This is supported by the fact that a vibration with a larger peak particle velocity (PPV) generates larger disturbance in rock and so finer fragmentation or more blast damage than a low PPV. The crack distribution in rock from impact is based on findings from previous researchers (Seaman, Curran and Shockey, 1976; Liu and Xu, in press). The fragmentation size is calculated at three-dimensional grid points within a blast. The fines and oversized blocks are calculated explicitly – the fines generated close to blastholes and big blocks far from boreholes are modelled naturally. The multiple blasthole fragmentation (MBF) model takes surveyed irregular geometry of the free face of a blast as the calculation boundary. The MBF can model fragmentation with over one million grid points. Examples of the model’s field comparison are also included.The PPV at a calculation grid is estimated based on the same approach used for the near field non-linear vibration model (Yang and Scovira, 2007). The method is based on the dominant charge for a calculation grid point, non-linear charge weight superposition within a charge and from multiple charges, strain wave broadening, and amplitude attenuation with distance from a charge segment. The blasthole confinement affected by earlier firing charges is also taken into account.The MBF model inputs from blast design, such as: location and orientation of each blasthole, stemming length, blasthole diameter, multiple decking, bench height, initiation sequence, etc. It models the effect of the delay and its scatter in each blasthole on fragmentation and has a statistical modelling capability for geological random variation on attenuation of particle velocity wave. The wave reinforcement due to instantaneous waves arriving or diminishing cooperative contribution due to long delay intervals between charges is also simulated.The results are displayed in 3D volumetric plots, contour of fragmentation in cross-sections, and size passing curves for a whole blast or a region of the blast.CITATION:Yang, R, 2015. A multiple blasthole fragmentation model – its concept, formulation, capability and field comparison examples, in Proceedings 11th International Symposium on Rock Fragmentation by Blasting, pp 177–186 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Aug 24, 2015

By Thomas L. Barkley

This paper describes an unusual design for a 12 foot diameter, 30 foot deep circular shaft shot in a close in urban environment less than 100 feet from occupied structures. It uses diagrams, video footage, and vibration results to describe a 3 deck, 17 hole burn round used to pull 17 feet in one lift.

Jan 1, 1994

By James R. Means, George Bowser

A new phone system was developed that provides eight full duplex communications channels over a single twisted shielded phone line, with no central switching station or line amplifiers. This system greatly simplifies installation and maintenance over that required with conventional switching systems. The system is based on microprocessor technology and digital frequency synthesis. Frequency division multiplexing is used over a bandwidth of approximately 300 kHz, and a baseband voice channel (compatible with conventional pager phones) is provided as a backup to assure communication if the multiplex features fail. Multiplex operations are controlled over a 30-kHz digital channel. The system is designed to operate with up to 16 km (10 miles) of cable plant and a maximum phone separation of 8 km (5 miles). The digital techniques permit a limited amount of monitoring and control capability to each phone. This consists of two pair of contact inputs or outputs, to be used as desired. These "flags" may be read or controlled by a dispatcher's phone.

By J Walshe, G Duclaux, L A. Fisher, R Hough, J S. Cleverley, M F. Gazley

Characterisation of hydrothermal gold systems across multiple scales is increasingly important for our understanding of ore deposit genesis and for targeting new mineralisation. Published work on multiscale mineral system exploration focuses either on applying a single technique across scales or on incorporating a series of techniques across a limited scale range. Typically, these studiesÆ spatial scales vary from crustal to deposit-scale and have successfully linked the distribution and shape of major orebodies to fault systems or varying lithostratigraphy. But, is this scale small enough to answer fundamental questions about the relative timing, processes interactions and mechanisms involved in the formation of ore deposits? We illustrate the potential of a deposit to microscale study, using the Plutonic Gold Mine in Western Australia as an example, to better understand hydrothermal gold systems. Using an integrated approach combining lithological and alteration logs, various geochemical data set, meso- to microscale structural and textural observations and spatial data assimilation, we have revisited the fundamental processes involved in the formation of this significant gold deposit.CITATION:Duclaux, G, Hough, R, Gazley, M F, Fisher, L A, Walshe, J and Cleverley, J S, 2013. A multiscale approach to hydrothermal gold systems - unravelling mineralisation processes at the Plutonic Gold Mine, in Proceedings World Gold 2013 , pp 95-98 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 26, 2013

By Roberto Carranza

The planning of experiments and the importance of statistical interpretation of the results have been recognized for years by scientists and engineers. The Factorial Design of experiments is among the various methods used for this purpose and is being widely applied in all branches of research and production. The basic principles of this tool have been described in detail (1, 2, 3, 4) and several examples of its application in various fields have been published (5). In recent years the mineral industry has also used Factorial Design to improve the efficiency of research, testwork and plant operation (6, 7, 8, 9). This method permits for a qualitative determination of the influence of the variables studied in a process and also a qualitative measure of the interaction of the variables. By graphic means or derivative methods, one may also establish the optimum conditions of the process (ibid 7, 8, 9). It is desirable, however, to develop mathematical methods to evaluate quantitatively these effects and restrict the optimization to the practical limitations of the variables. It is also convenient to adopt a notation system to generalize the Factorial Design.

Jan 1, 1971