Search Documents

By C McMahon, S G. Walters

Mineral maps of æintactÆ and large particle textures have been produced routinely from optical microscope images. Polished thin sections of large particles (1 - 3 mm) provide a balance between representative sampling and the preservation of primary rock textures. Optical microscopy is an efficient way to generate mineral maps suitable for geometallurgical applications. The automated optical mineral identification was carried out using a Leica DM6000¬ microscope with a high precision stage suitable for direct tiling of image frames and good registration of multiple image layers. Standard images collected were 10 x 10 frames (20 mm x 16 mm). Four image layers were recorded: ætransmitted light plane polarisedÆ, ætransmitted light crossed polarsÆ, ætransmitted light crossed polars with 1l tint plateÆ and æreflected lightÆ. All image analysis was carried out using Definiens Professional V5¬. The four image bands were read directly in 12 greyscale bands. A segmentation algorithm produced objects based on homogeneity across all bands. These objects were then classified using a complex set of rules (Process Tree) that are based on image brightness in all 12 bands. Opaque and non-opaque minerals were classified in the same sample. Using this protocol up to 12 minerals can be satisfactorily classified in one sample. The opaque minerals were recognised using colour ratio and reflectance intensity in the reflected light image. Non-opaque minerals are more difficult because of the complex variable colours of many minerals and the complexity of interference colours. However, object-oriented classification using three transmitted light image layers and multiple rules is a powerful technique for recognition of these minerals.

Jan 1, 2008

By S Telford, J Weir

Heathgate Resources Pty Ltd (Heathgate) operates the Beverley Uranium Mine in South Australia, located on the plains between the northern Flinders Ranges and Lake Frome some 600 km north of the state capital Adelaide. The uranium orebody exists in permeable sands within the confined, highly saline Beverley aquifer, approximately 125 m below the ground surface and the uranium is mined using in situ recovery (ISR) technology. The Beverley mine is a highly technologically advanced ISR mine and is operated in accordance with stringent safety and environmental standards. The mine is currently licensed to export 1500 t of uranium oxide (U3O8) per annum. This paper outlines the fundamental differences between ISR operations and the more traditional uranium production methods, such as agitation leaching and heap leaching, where the ore is removed from the deposit, stockpiled, crushed, milled and leached on the surface. Solids tailings from either the mill or the heap are finally deposited in a tailing storage facility. ISR mining differs from conventional mining because the ore remains in place underground (in situ) and the groundwater contained in the ore-bearing aquifer is utilised as the mining fluid. This is achieved by drilling into the deposit and installing wells for extracting and re-injecting the groundwater which is fortified with chemicals. The fortified groundwater (ie mining fluid) flows through the permeable sands containing the orebody, dissolving the uranium. The pregnant mining fluid is then collected and transferred to the recovery plant where the uranium is captured (on ion exchange (IX) resin in the case of Beverley). The barren mining fluid is then re-fortified and recycled to the wellfields. Once mining is completed in a wellfield, the surface infrastructure is relocated to a new wellfield, and the mined-out wellfield is rehabilitated to its original surface condition. A comparison of 'standard' hydrometallurgical plants, heap leach and ISR shows a number of key differences, such as the rate of leaching, the level of control available to the operators, and the type of contact/reaction that occurs. While limited in its applications, ISR mining has proven successful at Beverley and provides another mining and processing method for the production of uranium.

Jan 1, 2008

By A G. L Pratt, G F. Dunstan, A Leonforte

The selection of the material sizing and mine dewatering systems are important elements in the development of a large-scale, high tonnage (greater than 4 Mt/a) underground hard rock mine. The sizing system will influence the size and type of the ore haulage system and therefore how the underground project is developed. With more underground developments considering or implementing caving methods of mining, the selection of the material sizing systems is also influenced by the caving fragmentation. Traditionally, crusher systems are used for material sizing in large hard rock mines, with the crusher installation and its ancillary equipment representing a substantial capital investment. Underground cave mining type operations generally express themselves with large surface subsidence areas, which can lead to large inflows of water from major rain events or purging underground aquifer systems. To minimise disruption to mine production, the dewatering system needs to be sized, with consideration to both storage and pumping capacities. The balance between storage and pumping capacity is assessed to provide the optimal economic solution with the appropriate risk measures. This paper outlines a process for the selection of material sizing and dewatering systems for large-scale caving operations. This process is illustrated with information drawn from NewcrestÆs Ridgeway Deeps Block Cave project. The Ridgeway Deeps Block Cave is located directly beneath the existing Ridgeway Sublevel Cave Mine.

Jan 1, 2008

By P J. Hansen, D G. Franklin, M E. Trim, J V. Myers

The Northern Sewerage Project (NSP) is an upgrade to one of MelbourneÆs ageing, under-capacity trunk sewers. It will reduce the volume and frequency of wet-weather overflows to the Merri and Moonee Ponds Creeks and facilitate new housing developments in the northern suburbs of the city. A total of 12.4 km of tunnel and eight vertical access shafts will be excavated as part of the two project stages: Stage 1 (NSP1) will be owned and operated by Melbourne Water (MW) and Stage 2 (NSP2) will be owned and operated by Yarra Valley Water (YVW). The owners require the project to have a service life of 100 years, a requirement significantly beyond many of the current Australian Standards, which aim to achieve service lives greater than 60 years but are typically assumed to achieve 40 to 60 years. Because of the potentially corrosive natures of the external environment along the tunnels and the internal sewer environment, durability was a key concern for NSP1 and NSP2. Stringent specifications were applicable to the tunnel lining systems. This paper describes the corrosive environments impacting NSP1 and NSP2, explains the key design differences between the two stages and details the lining systems considered and selected to achieve the clientsÆ desired 100-year service life. The project design is complete and construction began in 2007.

Jan 1, 2008

Copper-Cobalt hydrometallurgy has seen a resurgence in recent years with the development of projects in Zambia and the DRC. Many of the flow sheets are unique in their use of hydrometallurgical techniques to improve performance compared to the older types in current use. The emphasis has moved from direct copper electrowinning/stripping, with multiple stages of precipitation for removal of impurities and concentration of cobalt, to innovative use of solvent extraction and ion exchange with higher cobalt recoveries to final product. The poor quality direct electro-won copper has been supplanted with solvent extraction-electrowinning; with consequent production of LME 'A' quality metal. Many of the precipitation steps in cobalt recovery and concentration, have been replaced with solvent extraction; while the remaining precipitation stages have been enhanced with use of oxidants to improve impurity removal. Operating experience from two of these circuits has shown the benefits of the new technologies and the improved overall recovery of both copper and cobalt into higher quality products.

Jan 1, 2008

By D Purdy, K Wilkinson, P Blake, M Fitzell, J McLellan, L Feltrin, O Dixon

The northern Drummond and Bowen basins region represents one of AustraliaÆs most prospective areas for epithermal vein and stockwork style gold deposits, with both low and high sulfidation epithermal mineralisation known in the region. 3D-GIS reconstructions, numerical and probabilistic modelling (UDEC, FLAC 3D and WofE) are used here to develop a better understanding of the controls operating on the mineral systems of the region and are aimed at enhancing predictive capabilities. An EXTENDED ABSTRACT is available for download. A full-length paper was not prepared for this presentation.

Jan 1, 2008

By A G. L Pratt, C Rossetti

The implementation of mass mining methods for large metalliferous deposits has led to the development and application of increasingly sophisticated automation and condition monitoring technologies in these mines. This technology aims to increase safety, equipment utilisation and monitoring of critical aspects of the underground ambient and geological environments. The communications networks required to support these aims and their associated technologies have evolved accordingly, and have done so in a demanding underground environment. The evolution of communication infrastructure for large underground mines has developed from basic analogue telephony to high-bandwidth optic fibre based systems. The latter is now the media of choice. More recently the developments in wireless broadband equipment designed specifically for underground environments have emerged to satisfy the needs of automation for mobile equipment and shaft conveyance communications. The use of internet protocols for industrial control and monitoring has led to the application of ethernet-based backbone networks as a standardised approach. There may also be a need for the integration of range of communications platforms in a mining operation to realise the full advantages offered by automation and condition monitoring technologies. As a result the design of the communications backbone networks for underground mines has started to become the domain of information technology professionals. Without these networks, the analysis of equipment utilisation is often not possible. The improvements that can be brought about by æsweating the assetsÆ and the transmission of business critical information are the most significant benefits of an integrated communications system. These benefits far outweigh the costs of strategic planning, subsequent specification and implementation of communication networks themselves. Future developments in mining systems and communications technology will continue to drive the demands placed upon communications networks as an essential mass mining infrastructure item. This paper outlines the benefits of secure high bandwidth backbone communications networks for mass mining operations and the need for a high degree of strategic thought and demand forecasting in the planning stages for these operations.

Jan 1, 2008

By S La Brooy, C Fountain

Throughout its history the mining industry has focused on the highest-grade deposits sited in the most convenient locations. The burgeoning global population and higher standards of living not only increase the demand for mineral products but also make it harder to mine and process them. The future will include the development of lower-grade, more complex orebodies, and orebodies in difficult locations. The industry will be expected to provide for humanityÆs needs while reducing its environmental footprint and maintaining profitability. This paper reviews some of the likely adaptations in mineral processing that will occur over the next 20 years. The challenges inherent in adapting to lower grade and more complex materials might be exacerbated by such factors as the availability of water and low-cost energy. In underground mining, higher energy costs are likely to foster further developments in underground preconcentration, especially for ores that can be upgraded at relatively coarse sizes. The adoption of underground crushing could also render material amenable to semi-dry preconcentration options such as might be achieved by the rotary classifier developed by CSIRO and being commercialised by RCR Tomlinson, or more modern versions of the hand-sorting of ore that use sensors based on colour or other suitable rock properties. The cement industry and some gold roasters have used the alternative of dry grinding and classification, which might also have a part to play in dry processing. There is also likely to be increased interest in options such as in situ leaching that further minimise the surface disturbance. Underground operations will have to balance the energy savings from reducing the amount of material brought to the surface and reduced surface disturbance against the practical limitations of installing and maintaining processing equipment underground. On the other hand, surface disturbance and energy issues in open cut operations will be balanced by potentially lower mining costs and higher overall recovery from the deposit.

Jan 1, 2008

By D Laurence

Significant developments in future mining systems, technology and best practices are emerging that are driven by issues relating to human resources, competitiveness, environmental and social responsibility, within the dynamics of globalisation. It is proposed that the definition of the role and attributes of the mining engineer in the mine life cycle needs to be viewed more holistically in the face of increasingly more diverse and complex responsibilities. These developments are motivating the need to consider a fresh approach to mining education and training at all levels for industry. A growing diversity and complexity of professional responsibilities is extending the mining engineerÆs broadly based skill set and reinforcing the need to play a central role within increasingly interdisciplinary teams. On the other hand, the imminent new technologies that will transform the mining process are prompting the assumption that more focused and specialised skill sets will be needed. A way forward is suggested that would first define a blueprint for the future mine, as an attempt to characterise the systems architecture, processes and organisation of the next generation mine. This can be the basis for assessing the nature of the workforce and distribution of responsibilities, from which can be inferred the skills and training requirements. In the context of the mining engineer, the paper explores how understanding such future requirements will enable more proactive development of a strategy for educational development. The paper then concludes by considering issues and lessons learned from some recent mining educational development initiatives underway in Australia and Canada that are shaping significant change.

Jan 1, 2008

By T E. Jeffries, M Zentilli, J STguel, C C. Wilkinson, B G. Rusk, V H. Vry

Laser ablation Inductively Coupled Plasma Mass Spectrometry (ICPMS) analysis of fluid inclusions trapping mineralising solutions in major porphyry-copper deposits indicates that copper concentrations can reach 103 - 104 ppm in both dilute, CO2-bearing fluids and in high salinity brines. Non-Cl transport of Cu is implicated. Au has been determined at 0.2 - 0.4 ppm in the low salinity fluids and may be co-transported with Cu. In contrast, high concentrations of Fe, Mn, Zn and Pb are only attained in brines and correlate strongly with Cl-indicating chloride transport. Physical and/or temporal separation of these fluid types can therefore producemetal fractionation within porphyry systems. Evidence concerning the behaviour of Mo is less clear but brine transport appears to be favoured; phase separation is one mechanism that may account for the fractionation of Mo from Cu and the formation of quartz-molybdenite veins. An EXTENDED ABSTRACT is available for download. A full-length paper was not prepared for this presentation.

Jan 1, 2008

By P Vasconcelos, X D. Deng

Numerous supergene manganese oxide deposits are developed inSouth China as a result of weathering of a thick sequence of Paleozoic marine Mn-bearing sedimentary rocks, consisting mainly of limestones, cherts and black shales that were intensely folded and faulted due to multiple episodes of tectonism since the late Triassic. The ore minerals are dominantly pyrolusite, cryptomelane, nsutite, lithiophorite and braunite. The supergene manganese ores contain Mn ranging from 20 to 44 wt per cent; manganese oxide deposits derived from cherts and black shales also contain recoverable levels of Co, Ni, Cu, Pb and Zn. 40Ar/39Ar dating of potassium-bearing manganese oxides reveals that extensive supergene manganese ore formation started at least in the earliest Miocene and persisted to the Present. Ages of the deposits show a close correlation to tectonic stability, with the oldest in tectonically quiescent areas and the youngest in active areas that have subjected to continuous uplift and thus significant denudation. The prolonged wet and hot climate since Oligocene and the balance between weathering and erosion, which is parameter to the tectonic stability, are the ultimate controls on the formation and distribution of South China supergene manganese oxide deposits. An EXTENDED ABSTRACT is available for download. A full-length paper was not prepared for this presentation.

Jan 1, 2008

I began my interest in sample preparation when I joined the Institute of Nuclear Sciences in 1962. I worked in the fields of rock dating and stable isotope geochemistry. Preparing a sample could take several days and the analysis took another day or more, so great care was taken with each sample. In 1968 I changed to mineral exploration analysis and was surprised to find that there was much less care in how a sample was taken and prepared. Virtually every sample was processed in the same way. In the use of ring and puck mills, Australia and New Zealand were among the world leaders. It took another ten years before labs in North America changed over from disc mills. In 1975, my partner and I divided our business into two parts. He took the laboratory into his consulting laboratory business and I took the small business that we had started, making ring and puck mills. For the next six years I was RocklabsÆs only employee. In the 1980s and 1990s, sampling specialists like Francis Pitard and Dominique Francois-Bongarcon began to make a difference in the understanding of what constitutes good sampling and preparation practice. They and others took courses on the new concepts of Sampling Theory, as developed by Pierre Gy. The desire for better procedures lead to the development of new machines, including the big bowl pulverisers from Australia, jaw crushers that could crush finely like our Boyd Crusher, rotating dividers of new designs and flow-through ring and puck mills. Other machines did not fare so well, including rolls crushers, cone crushers and hammer mills.

Jan 1, 2008

By T Raynlyn, J J. Campbell, R J. Holmes, J R. Manuel, S P. Suthers, A Poliakov

Prediction of the downstream processing performance of iron ore needs detailed information on mineral composition, textural peculiarities, particle density, porosity, mineral liberation and association and mineral grain size. Optical image analysis and automated iron ore texture classification systems are a way of obtaining such information. Furthermore, this information can be used for predicting the output of specific processes and their combinations. Several examples of the application of such an approach to process performance studies for iron ore are discussed. These include the application of classification by ore texture to modelling and optimisation of hydrocyclone performance, the effect of including textural information in modelling of sinter properties, and combining textural information with an experimental study of stirring and ultrasonic treatment of haematitic-goethitic iron ore fines to understand the deportment of fines and the breakdown/deagglomeration of particles. In all cases, the availability of textural information is important and helps provide a better prediction of process performance and/or understanding of the unit process.

Jan 1, 2008

By P M. Downes

Porphyry-related skarns are common around thePacific Rim. Examples include Cadia and Goonumbla (Australia) Ok Tedi (PNG) and Ertsberg and Grasberg (Indonesia). Although usually treated as separate to the adjacent porphyry copper-gold (Cu-Au) deposit, skarns near porphyry deposits are commonly inextricably part of the larger Cu-Au system(s). Using Cadia as the primary example, we demonstrate that large, zoned, calc-silicate skarns necessitate very large porphyry systems. The Big Cadia and Little Cadia Fe-Cu-Au skarns are associated with the Late Ordovician to Early Silurian high-K calc-alkaline to shoshonitic Cadia Intrusive Complex (CIC). Together, these skarns total ~40 Mt of Cu-Au ore but are much smaller than the adjacent Cadia EastûFar East, Cadia Hill, and Ridgeway porphyry-related Cu-Au deposits. Historically these skarns were large, economically significant and highly controversial deposits. Porphyry-related skarns typically have well-developed calc-silicate and ore zonation. At Cadia, classic skarn zonation has been mapped between Cadia Quarry (proximal) and Big Cadia (distal). Over this 800 minterval, mineralogy trends from: proximal garnet > pyroxene, grading to; fe-oxide-rich skarn; and distal haematite-epidote-rich skarn. At Little Cadia, analogous but symmetrical alteration zonation is present. Big Cadia formed from fluids that migrated laterally within calcareous units and were sourced from strongly altered quartz monzonite phases of the CIC at Cadia Quarry. Isotope studies and mass-balance calculations suggest that prograde and retrograde skarn alteration was caused by magmatic dominated fluids. They also indicate that the mapped alteration system at Cadia Quarry is insufficient to account for the total volume of Fe present at Big Cadia. Despite the typically oxidised nature of many porphyry systems, not all skarns in the porphyry environment are oxidised. Assimilation of carbonaceous sediment by hydrous, alkaline intrusions could lead to reduced Au skarns rather than porphyry deposits. An EXTENDED ABSTRACT is available for download. A full-length paper was not prepared for this presentation.

Jan 1, 2008

There are many different challenges when sampling mineral commodities. Sampling is part of many decision-making processes, from control of the number of drops of a reagent to all aspects involved in determining how much to pay for a mineral consignment. However, if all decisions must be supported by correct assumptions, or at least agreed terms, how can this be guaranteed? The main approach is to emphasise the importance of continuous surveying of sampling practices. If they are based on non-recommended protocols, this raises concerns. In addition, safety failures and lack of appropriate behaviour from the technical perspective can lead to disasters, both from the human and business perspectives. Impacts on safety and health are clearly unacceptable. It should be noted that imperfect safety and health practices in sampling operations have a strange correlation with incorrect sampling techniques. The commercial interface can also be affected. Companies have studied extensively the various aspects of measuring quality parameters for mineral commodities to understand the risks for investors and owners, and they take decisions that fundamentally respect the interests of parties in contracts and not result in losses. Every result should be obtained in defensible ways, and neither losses nor profits taken in indefensible ways. Most of such situations are related to bias; a significant systematic difference between a measured value and a reference value for a contractual quality parameter. Bias is always undesirable, because it always implies an incorrect advantage for one party and without doubt an undesired disadvantage for the other party. These situations are solved financially through a third-party arbitrator becoming involved in delicate situations, such as a commercial claim, but the outcome is always tangibly or intangibly bad for one of the parties. Another possibility is a permanent or temporary loss of precision. Poor precision, due its nature, generates reasonably balanced losses and profits, but it may hide out-of-control situations, generate instability and induce financial losses due to actions taken towards clearing commercial disputes involving interlaboratory programs, sample exchanges and duplicated sampling and sample preparation procedures, in the end wasting time and resources and creating a permanent source of commercial tension. While bias situations demand immediate adjustments for correctness, poor precision has to be understood as something that cannot be just calibrated out and is a situation that demands investment of time and effort. Precision optimisation has to take into account best practice for obtaining the required precision as well as cost/benefit analysis studies. Purchasers, engineers, salesmen and all personnel need to recognise that sampling is a major area to understand when taking decisions. The aim is to spread concepts that can really help, such as simple in-house actions to improve productivity and completion of surveys on commercial sampling operations, in order to ensure correct results are obtained and assertively take corrective actions, thereby establishing a healthy relationship with investors, suppliers and purchasers and in the process optimising the business.

Jan 1, 2008

By D A. Collins

The floatability methodology developed by the AMIRA P9 project through research contributions by the Julius Kruttschnitt Minerals Research Centre (JKMRC, Brisbane, Australia), University of Cape Town (UCT, Cape Town, South Africa) and McGill University (Montreal, Canada) has been successfully commercialised through JKTech. The methodology involves detailed flotation plant campaigns which require large quantities of data to fit models that can be used in JKSimFloat (a steady state simulator for the flotation process). The methodology to date, although widely accepted as the best flotation optimisation technology available, is limited to existing circuit optimisation and requires full plant or pilot plant data to calibrate new models. The floatability of new ore types cannot be characterised using this method unless a pilot campaign is undertaken. After reviewing a large database of existing industrial floatability data, a new methodology is proposed that involves a laboratory based flotation test procedure and review of the floatability database to characterise the floatability of new ores. The full methodology is labelled the floatability index test procedure (JKFIT). This paper outlines the JKFIT procedure and the results of applying this methodology to pilot plant data from a copper ore. JKFIT was applied to two different feed sizes (namely a P80 of 180 microns and 212 microns) of this copper ore. The results of applying JKFIT to this case showed that the metallurgical response (in terms of grade and recovery) from the full pilot plant survey data could be predicted well. Further validation of this methodology was conducted on a separate copper ore. The results from this validation also showed that the JKFIT test was able to be used to predict the full plant performance. The JKFIT test has the potential to become the 'drop weight' test of flotation which will improve the confidence in plant design flow sheet layouts, allow full orebody characterisation and prediction of key grades and recoveries before the ore enters the plant, and will provide an improved methodology for chemistry impact analysis on different ores.

Jan 1, 2008

By R Berry, N Bonnici, K Nguyen, S G. Walters, M Kamenetsky, C L. Evans

Mineral maps produced via optical and scanning electron microscope û mineral liberation analyser (SEM-MLA) automated mineral identification systems are being used to examine the relationships between mineral phases in spatially well-constrained samples of intact and crushed rock. The use of sophisticated image analysis software allows data pertaining to individual or groups of mineral phases to be extracted from the mineral maps. Analyses of the extracted data can be used to identify populations of phases. Data extracted from the images can be compared with mineral processing data, eg hardness, grindability, degree of liberation, from the same sampling intervals, enabling ore characterisation to be linked to processing behaviours in a spatially constrained, quantified manner that can potentially assist with mine planning and optimisation.

Jan 1, 2008

By J-H Shin, Y-S Jung, D-J Youn, G-J Bae

This research is to grope for a method to reduce high pore pressure taking place on tunnel shotcrete lining, especially in underwater tunnels, effectively by utilising a new concept of patternised pin drain. Accordingly, the patternised pin drain system is investigated numerically in order to confirm to the effect of reduction of pore pressure by the pin drain and its effective application. Installation of pin drain shows considerable reduction in the figure of pore pressure. It is found through 2D numerical modelling that the change of pore pressure, according to the shape and the location of pin drain, is sensitive by the sequences of the change of location > the change of length > the change of angle degree. In the results of investigation in 3D modelling, the pore pressure is the largest in the centre of the interval among pin drains, and it is possible to obtain an ideal degree of longitudinal angle and an interval by investigating pore pressure taking place on the lining.

Jan 1, 2008

By G Gibson

Realising that current roadway development performance levels were unlikely to sustain new generation high-capacity (15 Mt/a) longwall mines then being planned, the Australian Coal Association Research Program (ACARP) established a Roadway Development Task Group (RDTG) in 2005 to develop a research strategy targeted at improving roadway development performance in Australian longwall mines. The initial phase of the project identified a number of factors that differentiated best practice mines from others, and proposed a multi-faceted research strategy encompassing both short- and long-term research initiatives, including:conducting regular benchmarking studies across the industry to identify factors contributing to improved performance and to monitor performance trends,conducting regular industry roadway development operatorsÆ workshops to facilitate the transfer of successful roadway development improvement initiatives throughout the industry and facilitate networking between development practitioners,development of alternative skin reinforcement and containment measures that would eliminate the use of steel mesh and enable face support systems to be fully automated,development of automated bolting and mesh handling systems that would enable operators to be removed from the immediate face area,development of a body of knowledge of industry roadway development practice,development of a web-based information management system to capture and transfer emerging developments in roadway development practice and technology,reviewing the Australian civil tunnelling and underground metalliferous mining sectors to identify practices and technologies that could potentially be utilised to improve roadway development performance, anddeveloping a new generation high-capacity mining system capable of +10 metres per operating hour for 20 or more hours per day (CM2010). The paper reports on the progress of the above initiatives and outlines strategies to continue the pursuit for improved roadway development performance.

Jan 1, 2008

Operating metallurgists spend a significant part of their time running plant trials to compare two or more conditions in order to improve the process, such as a new reagent, flow sheet or piece of equipment. The best way to conduct such a trial is as a paired experiment when testing two conditions (eg 'old' and 'new') or a randomised block design when testing more than two conditions. These experimental designs should always be preferred over alternatives if their requirements can be met, because they block the effect of uncontrolled variables ('covariates') and are the most statistically efficient, ie they will lead to the required level of confidence in the outcome faster than other methods. Sometimes, however, it is not possible to switch condition easily, which is a prime requirement of these methods. In such cases the only option may be to switch condition once only, and compare the process performance before and after the switch. Examples are testing grinding media, large equipment, or long residence time processes such as leach trains or CCD thickeners. This paper considers some data analysis methods to deal with this situation. Modelling with intervention analysis (IA) uses regression models with a dummy variable to indicate the presence or absence of the test condition. IA with a time series model ensures that the residuals are uncorrelated. IA with a process model helps to deal with the problem of covariates. The reference distribution is a method free of any statistical assumptions but is less powerful than the other methods. Cusum charts are a helpful visual aid to interpreting time trends. A combination of these methods can help improve the confidence in the final decision. None are as good as a formal experimental design such as a paired trial.

Jan 1, 2008