Search Documents

By J. A. Hudson

"The paper begins with a brief summary of Professor Müller’s formation of the ISRM and the contents of the previous six Müller lectures with their emphases on rock fractures. The geological origin of fractures is described via a deformational spectrum diagram and a matrix characterising the consequences of a sequence of superimposed deformational events. The different fracture frequencies encountered along lines in all directions through a fractured rock mass are explained. The value of computer modelling and simulation is demonstrated through an example of the effects of fracture zones on the distribution of the in situ stress field. The need to establish the validity of computer programs, i.e., that the modelling/simulation, does actually represent the rock reality, is discussed with reference to the work of the DECOVALEX project. Finally, examples are given of the effects of rock fractures on engineering design and construction, from pre-split blasting to caverns.INTRODUCTIONIt is a great honour to have been chosen as the recipient of the ISRM 7th Müller Award and I am grateful to the representatives of the International Society for Rock Mechanics (ISRM) National Groups who voted for me. In addition, I am especially pleased because my working life in rock mechanics and rock engineering has been devoted to understanding the geometry and mechanical properties of rock fractures and rock masses—which were the key subjects that Professor Müller promoted. In the ISRM 50th Anniversary Commemorative Book, Hudson and Lamas (2012), Professor Charles Fairhurst notes that on 24 May 1962, i.e., the day before the constitutional meeting of the ISRM took place in Salzburg, a reporter asked Professor Müller, “Do we know the strength of rock?”, to which Professor Müller replied, “For rock tested in the laboratory, yes. For a rock mass, no. This is what we need to determine. This is why we need an International Society for Rock Mechanics”. Also, in his Opening Address at the 1st ISRM Congress in Lisbon, Professor Müller, as the then ISRM President stated,“Many experts agree with me that discontinuity and anisotropy are the most characteristic properties of the material rock and that the properties of jointed media depend much more upon the fabrics bond of the unit rock block system than upon the rock material. Therefore any theoretical investigation of that material has to go its own ways, in the same way as the construction material of soils years ago suggested to soil mechanics its own methods, which differ greatly from the way of thinking of technical mechanics”."

Jan 1, 2015

By G. Goodall, D. Mercier, M. Brochu, C. Kaplin

A nanostructured CoNiCrAlY coating was obtained by thermally spraying cryomilled feedstock powder using the high-velocity oxy-fuel (HVOF) process. The oxidation behavior of the nanostructured powder and coating were compared to their conventional counterparts by performing static oxidation tests at 1000°C for durations ranging from 1 to 96 hours. It has been found that oxidation of the conventional powder yielded a duplex oxide scale composed of an inner layer of Al2O3 and an outer layer composed of Cr2O3, NiAl2O4 and other spinels. Interestingly, oxidation of the as-sprayed nanostructured coating yielded a similar duplex oxide scale while oxidation of the polished coating resulted in a single Al2O3 layer. These findings indicated that surface oxide seeds act as nucleation sites for the growth of the mixed oxide layer. Also, it was determined that nanostructuring of the coating promotes the nucleation and growth of a single and adherent a-Al2O3 layer by increasing both the number of diffusion channels and the aluminum concentration at grain boundaries.

Jan 1, 2010

By T. A. Utigard

To improve recoveries of Ni, Cu and Co in the Inco flash furnace, a new operating concept was investigated. By adding sufficient coke to form a supernatant coke layer, a reducing barrier would form allowing for improved reduction of descending oxidized sulfide mineral particles formed in the flame. The objective of this investigation was to study the kinetic mechanism and efficiency of the coke layer to produce a sulfur deficient matte. The mechanism and rate of carbon reduction of each of the phases formed in the flash flame, i.e. sulfides and oxysulfides were investigated by means of X-ray visualization tests. Graphite-slag reduction was also conducted for comparison purposes. In the case of slag, poor wettability and the formation of a gas film on the surface of the carbon reductant are often blamed for limiting the reaction surface area that ultimately results in poor reaction kinetics. The X-ray visualization tests confirmed that matte and oxysulfide wet coke better than slag and that the rates of reduction of matte and oxysulfide are faster than that of slag.

Jan 1, 2005

By J. S. Edwards

The ISASMELT? process is recognized worldwide as a modem, flexible, environmentally-friendly and low capital cost smelting technology. Due to the inherent simplicity, and the advanced process control that has been developed, personnel become competent within weeks of being introduced to the ISASMELT? technology. This compares with the years required to become proficient with alternative technologies. Copper smelters in Australia, USA, India, China, Peru and Zambia have installed ISASMELT? technology under licence from Xstrata Technology. In the secondary copper business, plants in Belgium and Germany are processing copper and/or lead materials in ISASMELT? furnaces in a two stage process. In a single furnace smelting is followed by converting. Both plants have exceeded design capacities and are continuously improving their performance. The next evolutionary step for the technology is the implementation of the continuous copper converting process on a commercial scale. The present work describes the design status of the continuous converting process - ISACONVERT?.

Jan 1, 2007

By B. Voyzelle

The deterioration of highway and bridge infrastructure has received considerable attention in recent years because of the billions of dollars needed for repair and rehabilitation. A major factor contributing to this problem is the corrosion of steel reinforcement. Reinforcing steels with good corrosion resistance and relatively low price are therefore essential for maintenance, cost reduction and service-life extension of the steel-reinforced structures. This article describes the development of low-alloying steels that offers good corrosion resistance with reasonably low cost. Topics include steel microstructures, mechanical properties, corrosion performance, industrial applications and economic aspects.

Jan 1, 2005

By W J. Riva

This paper reviews the circumstances that led to the development and construction of a commercial-scale vertical shaft carbonizer at Conmore, Alberta. This form coking process may be regarded as having potential application in other areas. Coal pelletizing offers a promising method of decreasing coal agglomeration costs under certain circumstances. Pellets can be rolled easily from -35-mesh coal, provided that a minimum of -200-mesh coal is present. Bentonite was found to be the best binder and was most effective when added in amounts of about 2 per cent. The water content of the coal pellets was observed to vary from 18 to 23 per cent. As a by-product of this investigation, it seems possible that fine coal could be converted into stoker feed by this process. A short investigation of the effect of heating rate on the carbonization of pellets was conducted. The low maximum heating rates in certain temperature ranges, necessary to avoid crack formation in the pellets as reported by other

Jan 1, 1964

By V. Maazallahi, A. Rouleau, A. Saeidi

"The geometry of the terrain topography before and after the failure are the essential factors for analyzing a landslide and for the design of remedial works. This paper presents a method to determine these factors, considering that the determination of the slide surface profile is difficult and costly using traditional methods and the pre-slide topography data is often not available. The presented method to locate the 3-dimensional geometry of the slide surface is based upon determining the slide surface in 2- dimensional vertical sections. This method utilizes post-slide topography of the area to determine the pre-slide topography and uses equations to estimate the pre-slide situation using a statistical process applied to points located on the sliding mass. This method is tested on the Maskun landslide in Iran. The results indicate the applicability of the proposed method to obtain the key factors for the back analysis of a landslide in a cost effective manner.INTRODUCTIONProbably the most reliable method of determining the strength of soil and rock mass is to back analyze a failed or failing slope (Caia et al. 2007; Wyllie & Mah, 2004). This method is used in geotechnics for estimating geotechnical parameters of the rock and soil mass (Sakurai et al., 2003; Frayssines and Hantz, 2009; Cepuritis et al. 2010), but it requires that the failure mode is well established and that there is complete and precise information available on the slide surface and the sliding mass (Sjöberg, 1996). Measurement of the slope profile before failure provides the basic information required for a back analysis (Hock & Bray, 1981). The post-slide topography is another key factor. The back analysis process for determining the actual strength parameters of soil or a decomposed rock mass after a landslide requires knowledge of the sliding mass geometry, both before and after the sliding event. Many studies are reported in the literature about the application of back analysis for determining geotechnical and geomechanical parameters of rock and soil mass (Sakurai et al., 2003; Cepuritis et al. 2010; Frayssines and Hantz, 2009, Hussain and Stark, 2011), but they generally present little information about the geometry of the slide surface, and the weight of rock and soil mass that slid, although these factors are the most important parameters for back analysis process. Liang and Xue-song (2012) determined the slip surface having the minimum factor of safety (called critical slip surface in deterministic analysis of slopes) and the slip surface with the minimum reliability index (called critical reliability slip surface), but they have not considered the actual profile and geometry of the slide surface. Rocscience (2013) proposes the back analysis option in the Slide software for slope stability study, but it requires an imposed slide surface as input; alternatively, one can estimate by a finite element analysis the surface with the minimum factor of safety. However, this estimated surface is very likely different from the actual sliding surface, considering the uncertainties in input data of the finite element method."

Jan 1, 2015

By F. Daniel Rosenkranz, Peter L. Silveston

Experiments investigating the treatment of a ground wood white water waste from an Ontario pulp and paper mill by an HVCB coal are reported. Adsorption capacity for the waste, with a soluble COD contribution of 2,100 mg/I, amounted to 0.17 per cent, by weight, of the Nova Scotia coal used. Percolation of the white water waste through a coal bed indicated that soluble COD is removed. For finely ground coals, over 90 per cent of the suspended solids in the waste are retained on the coal. Plugging of the coal bed is rapid, so that any treating process using filtration would have to renew the surface of the bed frequently.

Jan 1, 1970

By R. Pressacco

"IntroductionAgrium’s history in fertilizer production and marketing goes back to 1931. Today, Agrium Inc. is a world leader in the production, distribution, and marketing of fertilizers and related products to the agricultural industry. Agrium is the largest North American producer of nitrogen fertilizers, a major producer of phosphate and potash fertilizers, and a world leader in fertilizer wholesale distribution and retail marketing. The mine at Kapuskasing is located some 40 km southwest of Kapuskasing, Ontario (Fig. 1). It has been developed to supply high-quality, low-cost phosphate rock to the company’s Redwater, Alberta facility, where it is an ingredient in ammonium phosphate fertilizer production. The Kapuskasing mine is the only phosphate mine in Canada and will rank as one of the highest-grade phosphate mines in the world (Clark and Duncan, 1999). Apatite is the principal ore mineral mined at the Kapuskasing Phosphate Operations.GeologyThe first record of exploration in the area was from a program executed by Continental Copper Mines in 1955 which evaluated the magnetic anomaly located in the Cargill and Cumming townships for its base metal potential. Intermittent exploration of the base metal potential of this magnetic anomaly continued in the following years, until the phosphate potential of the area was first examined in 1974 by International Minerals Corporation which recognized the presence of a layer of apatite sands overlying the carbonatite host rocks.The local geology consists of a core complex of multi-phased carbonatite rocks which are surrounded by a ring of pyroxenite, and have provided a U-Pb age date of 1907 Ma ± 4 (Sage, 1988). These two rock types are in turn situated within quartz diorite gneisses that form a large portion of the Kapuskasing Structural Zone (Fig. 2). The carbonatite host rock is sub-divided into two sub-types: sovite and rauhaugite. The sovite is a medium- to coarse-grained, white, banded rock in which calcite is the dominant carbonate species and it includes accessory minerals such as phlogopite, magnetite, clinohumite, apatite, olivine, pyrrhotite, and amphibole. Apatite can reach 15% abundance in this rock type (Sage, 1988). In sharp contrast to the sovite, the rauhaugite appears as a massive, fine-grained, dense, beige to tan-coloured rock in which dolomite is the dominant carbonate species. Phosphate values can range to 14% P2O5 in the rauhaugite."

Jan 1, 2001

By J. S. Stachulak

"The main ventilation design features adopted at Inco to safeguard personnel against the effects of fires are described in this paper. Measures taken to prevent and detect fires due to spontaneous combustion, and fires involving conveyor belts, are described in detail.IntroductionConsidering the over-all picture of present-day mining operations, the expenditure required to provide a proper system of mine ventilation and fife control should be regarded as a sound investment that will pay back in increased mining efficiency, improved safety, health, and morale of the miners.Mine ventilation control and mine fife detection and fighting are inseparable, because a ventilation system can spread deadly products of combustion just as efficiently as it does life-sustaining oxygen. Therefore, the possibility of a mine fife occurring is of particular concern.The first objective is to prevent a fife.The second objective is to design the ventilation system so that a positive fresh air supply can be ensured on each main level regardless of the location of the fife.The third objective is to be able to detect the fife at an early stage and to extinguish it before it can develop into a major fife. Accordingly, at Inco, very extensive measures are taken to prevent fires and to locate, isolate and extinguish them rapidly. To obtain an early warning, it is not sufficient to know that a fife has started, but it is also necessary to know where the fife is burning."

Jan 1, 1994

By J. S. Stacbulak

"The main ventilation design features adopted at Inco to safeguard personnel against the effects of fires are described in this paper. Measures taken to prevent and detect fires due to spontaneous combustion, and fires involving conveyor belts, are described in detail.IntroductionConsidering the over-all picture of present-day mining operations, the expenditure required to provide a proper system of mine ventilation and fife control should be regarded as a sound investment that will pay back in increased mining efficiency, improved safety, health, and morale of the miners.Mine ventilation control and mine fife detection and fighting are inseparable, because a ventilation system can spread deadly products of combustion just as efficiently as it does life-sustaining oxygen. Therefore, the possibility of a mine fife occurring is of particular concern.The flirt objective is to prevent a fife. The second objective is to design the ventilation system so that a positive fresh air supply can be ensured on each main level regardless of the location of the fife. The third objective is to be able to detect the fife at an early stage and to extinguish it before it can develop into a major fife.Accordingly, at Inco, very extensive measures are taken to prevent fires and to locate, isolate and extinguish them rapidly. To obtain an early warning, it is not sufficient to know that a fife has started, but it is also necessary to know where the fife is burning."

Jan 1, 1994

By G. E. Agar, W. B. Kipkie, R. A. Tenbergen

"This paper describes the successful pilot plant testing of a flowsheet, developed at Inca Ltd.&apos;s J. Roy Gordon Research Laboratory (JRGRL), for improved separation of the constituent minerals in a bulk copper-nickel concentrate.The flowsheet utilized the fact that xanthate concentration on the mineral surfaces is the most important variable affecting the high lime separation and that reduction of the xanthate concentration could be affected by an oxidative conditioning stage.When the adsorbed xanthate was reduced to a suitably low level, chalcopyrite and pyrrhotite could be floated to leave a high-grade nickel concentrate as the residue. Sodium cyanide was added to the froth product 10 depress pyrrhotite and any remaining pentlandite. The temporary depression of chalcopyrite by the sodium cyanide was overcome by introducing a conditioning step following cyanide addition . The advantages of the flowsheet; improved copper-nickel separation, operation without heating of the pulp, production of both high-grade nickel concentrate (29.5% Ni, .26% Cu) and copper concentrate (<0.5% Ni, 30% Cu) were all realized.IntroductionLaboratory test work undertaken at Inco and described previously(1) has demonstrated the importance of low xanthate concentrations on the mineral surfaces in the copper-nickel separation at the Copper Cliff mill.It was determined that oxidation of the separator feed (a cleaned bulk concentrate, grading about 11% Cu and 11% Ni) to lower the xanthate concentration on the mineral surfaces prior to separation significantly improved the quality of the separation (CI 4)*, particularly when no cyanide is used, but also when cyanide is added. As well, it was found that the oxidation permitted a pentlandite-pyrrhotite separation - the oxidation step brought about a decrease in the flotation-rate of pentlandite leaving pyrrhotite and chalcopyrite largely unaffected. Consequently, it was possible to produce a highgrade pentlandite product (30% Ni, 0.3% Cu) at ~83% Ni recovery , as well as a high-grade copper concentrate (0.5% Ni, 30% Cu at -96% Cu recovery)."

Jan 1, 1985

By G M. Thorpe

This paper discusses mine research at Inco, giving specific examples of research projects which illustrate the approach to mining problems. The first example covers the development of a new mining method, "undercut and fill," which was originated for mining badly broken-up pillars, but which is now rapidly replacing the square set method of pillar recovery. In the new method, ore is extracted in a series of successive horizontal cuts, progressing downward, a cut being completely mined and filled before the cut below is started. In the second example, improvements to the "cut and fill" method of mining are discussed. These improvements are the result of improved techniques and improved equipment, etc., with, in many cases, the former being made possible by the latter. The advances made have permitted the use of the "cut and fill" method at much greater depths than before. The third example is presented to illustrate research undertaken to obtain the earliest maximum advantage of the introduction of a new piece of equipment or material. For this purpose, the development of the "Under-ground use of AN /FO Blasting Agent " has been selected. Work on this project started in May, 1960, at a time when there was very little in-formation regarding the potentialities of AN /FO for underground blasting, either primary or secondary. By the end of 1961, AN /FO made up 64 per cent of the explosives used under-ground at Inco mines.

Jan 1, 1963

By C. S. Lord

"AbstractThe slow progress of conventional geological reconnaissance of the northern Canadian Shield during the past twenty years clearly indicates the need for a radically new field technique. Operation Keewatin, involving the use of helicopters as the traversing vehicles, was planned and executed by the Geological Survey of Canada in response to this requirement.About 57,000 square miles of District of Keewatin, N.W.T., were mapped during the 1952 field season in appropriate detail for publication at 1 inch to 8 miles. The Operation disclosed about 14,100 square miles of Precambrian sedimentary and volcanic rocks ranging in •age from Archaean to Late Proterozoic. About 7,700 square miles of this area are underlain by greenstone. This work was accomplished in a field season of 113 days, including the spring break-up. Routine aerial geological observations were made from two helicopters flying about 300 feet above the ground; these observations were checked by more than 1,700 landings and about an equal number of aerial observations from heights of less than 50 feet. Logistical support was •provided by a Norseman aircraft on skiis or floats. Combined helicopter and Norseman flying time was about 990 hours. More than 1-00 square miles were mapped per helicopter-hour, and about 815 square miles were examined each day that both helicopters operated. The reconnaissance cost about $207,000, or $3.63 per square mile mapped.As a result of this trial use of helicopters there is now available a reasonably proven method whereby an acceptable geological reconnaissance of the barren grounds of the Canadian .Shield within the Districts of Keewatin and Mackenzie might be completed within a decade. Thus, within a few years, it could place at the disposal of the prospecting and mining fraternity data which, by conventional methods, might not be made available to this generation. Due weight must be accorded this time factor when evaluating the aerial technique."

Jan 1, 1953

By M. E. Holuszko

A research program has been carried out at the Mining Department of the University of British Columbia to study the effect of wettability of coal on its handleability and bulk density. The effect of moisture content on the bulk density of coal was found to be more pronounced for hydrophilic coals than for hydrophopbic ones. Altering wettability of coal surface leads to different effects of moisture on bulk density. The samples were studied at different fine fractions content, and it was shown that there was no effect of increased content of fines on the bulk density for hydrophobic coals, while for hydrophilic coals this effect was very significant. It was shown that the moisture content at which the lowest bulk density was reached coincides with the critical moisture content for studied samples, as determined from handleability tests using Durham Cone method. This critical moisture is the moisture content at which a sudden deterioration of coal handleability occurs. Based on the bulk density and handleability tests, the flow theory of moist coal is proposed.

Jan 1, 2006

By J. H. Morrish

"The delivered price of any mineral is the determining factor in converting untapped resources into economic resources. Transporting the minerals from the mine to the market plays a significant part in the total package. The expansion of the Western minerals industry in the last ten years has been paralleled by improved transportation systems to reduce cost, increase accessibility of resources and get the minerals to market in the shortest time possible. A review is made of developments, such as unit trains, complex multi-modal systems and containerized transport, over the last ten years, with specific movement outlined. The total transportation package of land transport, port and terminal facilities, and water transport, where needed, is stressed. As the volume of minerals movement expands, increasingly sophisticated and integrated transportation systems will have to be utilized. The development of solids pipelines and other possibilities is outlined."

Jan 1, 1973

By R. C. Gilmore, J. F. Brune

"Pulverized limestone rock dust used in United States underground coal mines to prevent and suppress coal dust explosions must easily disperse by forces of a methane or coal dust explosion. Moisture in the mine environment causes caking of the dust. Colorado School of Mines researchers performed dispersibility tests in a full-size explosion test drift. Tests included conventional, hydrophobized, and rock dust approved for German mines. Conventional and German standard rock dust products partially agglomerate, which may not be effective in suppressing a coal dust explosion, while hydrophobic rock dust retains better dispersibility.INTRODUCTIONA coal dust explosion devastated the Upper Big Branch mine in 2010 resulting in the death of 29 miners, the worst mining accident in the United States since 1972. The explosion propagated through approximately 65 km of active mine workings, but was extinguished in areas that had been properly rock dusted (Page, 2011). Researchers have long studied and established the ability of inert limestone rock dust to suppress coal dust explosions, see Richmond et al. (1975), Cybulski (1975), and Sapko et al. (2000).To be effective, rock dust must contribute to a total inert content of > 80% by weight and 100% of rock dust particles must pass through an 850 µm sieve, while 70% of the rock dust must be finer than 74 µm. This is established in US regulations 30 CFR §75.2. Dry rock dust is a nuisance that prevents mining work downwind from rock dusting activities. To overcome this problem, dust manufacturers have proposed applying the rock dust as a wet slurry or foam. After drying, manufacturers expected the slurry or foam-applied dust to have the same explosion suppression capabilities as dry-applied rock dust.Researchers at the Colorado School of Mines (CSM) designed and built a full-scale, underground Explosion Test Facility (ETF) at the CSM Edgar Experimental Mine in Idaho Springs, Colorado about 20 miles west of the CSM Golden, Colorado campus. To disperse the different rock dust products, researchers produced an air pressure pulse using a mesh array of PETN detonation cord. This method is simpler, more reproducible and safer than other methods considered, including methane-air explosions and a compressed air cannon. The detonation cord blasts produce a wind velocity of 30 to 50 m/s, the minimum velocity at which coal dust explosions can propagate, based on research by Nagy (1981) and Cybulski (1975)."

Jan 1, 2019

By Jürgen Kretschmann

The Islamic Republic of Iran is a very important producer of raw materials in the world. The country holds 68 types of minerals, including salt, sand and gravel, chrome, lead, zinc, copper, coal, gold, and iron-ore. With more than 37 billion tons of measured and indicated reserves and 55 billion tons of total mineral resources, Iran is ranked among the 15 major mineral-rich countries. Sangan iron-ore mines (SIOM) is one of the biggest iron ore mining complexes in Iran, located in a remote, but strategically important area near the Afghanistan border. The Iranian government has planned several mining projects in this region for a stepwise development over the next 5 years to satisfy the growing demand of the Iranian steel industry. In cooperation with private companies, these projects shall be realized as sustainable as possible. This paper introduces SIOM and describes socially responsible activities undertaken during the initial development of this mining complex. The effects of these activities will be demonstrated. Unique attributes of the Sangan area have a significant impact on the acceptance of SIOM by regional stakeholders. The government has made several social investments that benefit workers´ families and the community. They are attempts to build social capital for the mining operations and demonstrate a commitment for socially responsible mining to the communities and the local population. They aim to create a balance between the local and regional expectations and requirements and the necessity to secure the corporate profitability at a reasonable and sustainable level. Investment in socially responsible activities is not a choice but a necessity to improve the relations between mining companies and stakeholders in the communities. It is recommended that social investments should be carefully planned and implemented from the early stages of mining activities, and should proceed throughout the life cycle of the mines.

Aug 1, 2013

By R. C. Beard

"Heap leaching, currently being applied in the treatment of low-grade ores in the south western United States, is shown to have potential for use in north western Ontario. The history and operation of the process are described, and cost comparisons are given.IntroductionHeap leaching is a simple, low-cost method of recovering gold from low-grade ore. Although it is not currently being used in Canada on a production scale, it is being used extensively, and successfully, in the southwestern United States as well as elsewhere in the world. Especially suited for large deposits of very low -grade ore, it also appears to be practical for use on small, high-grade deposits because it avoids the high capital costs of mills. Eveleth (1980) and others suggest that the capital and operating costs required for heap leaching are approximately 20% and 40% respectively of conventional milling methods.Northwestern Ontario is noted for its large number of undeveloped, small, high-grade gold deposits. Although there are a number o f larger and higher-grade deposits in the area as well, it is the relatively small undeveloped deposits which would benefit most from heap leaching, if it could be demonstrated to be a viable extraction method.To assess the significance of these small deposits, the Ontario Ministry of Natural Resources recently commissioned a consultants&apos; study of the gold reserves in certain parts of northwestern Ontario. This report, titled ""Feasibility of Small-Scale Gold Mining in Northwestern Ontario"", by J.N. Neilson and R.C.E. Bray (1981, OGS Open File Report 5332), documents 106 known gold deposits having sufficient information to calculate gold reserves. A total of over 7.8 million tons having a weighted average grade of 0.19 oz of gold per ton were compiled in the proven, probable and possible categories, with an additional 8.2 million tons in the speculative category. Many of these deposits could be mined by open-pit methods."

Jan 1, 1983

By J. P. Harrison, N. Bozorgzadeh, M. Dolowy-Busch

"It is common in rock engineering practice to characterize the triaxial strength of intact rock by using only data samples of small size. Practical guidelines such as ASTM support such sample sizes. In this paper, we investigate the reliability of the Hoek-Brown strength criterion estimated from such samples. For this purpose, we use an unusually large published set of triaxial strength data to randomly select a large number of data samples of small size to simulate the situation that frequently encounters in practice. The samples are generated with different arrangements with regards to the confining pressure. We fit the Hoek-Brown criterion to each of these samples, obtain estimates of strength parameters and compare them to those obtained from the complete data set. Our analysis suggests that estimates based on small samples are not reliable and can lead to both over- and under-estimation of strength.Additionally, we use these small samples to investigate the statistical methods for obtaining design values of strength as suggested by limit state design standards such as Eurocode 7. We show that, if used in conjunction with small samples, these methods do not guarantee a safe estimate of strength and in many cases lead to highly over-conservative estimates.We present a potential solution to this problem by augmenting the small data sets with Bayesian informative priors to give much more reliable estimates of strength. We conclude with suggestions for the development of informative prior distributions."

Jan 1, 2015