Search Documents

By J. C. Hampton, L. A. Matzar, C. R. Gamwell

In geomechanics, conducting uniaxial or triaxial compression tests require samples to be prepared to a certain length-to-diameter ratio. This ratio is typically defined as 2:1 (length to diameter) and is accepted as a common standard for all materials on which mechanical properties testing is performed. Variations in sample length can cause changes in measured strength properties across multiple material types, such as cement, rock, or soil. Within the oil and gas industry, it is not uncommon to receive samples for technical services work that are slightly under or exactly the length requirement before surface grinding and sample preparation. Many of these samples are discarded from uniaxial and triaxial compression tests and used in another manner; therefore, it is necessary to study the length-to-diameter ratio in efforts to understand the limits and validity of testing samples that are slightly out of the 2:1 ratio. This paper discusses uniaxial compression tests performed using Berea sandstone and a standard cement mixture of 2.54-cm diameter to study changes in rock strength and properties. Compressive strength for each lengthto- diameter ratio over a range of 0.5:1 to 2.2:1 was measured. The investigation was extended to study near-end effects, where the sample length is near, but not exactly twice the diameter. It was discovered that the cement samples conformed reasonably well to the suggested ASTM correction factors. The sandstone did not conform to the correction factors and further analysis is required to develop a correction factor unique to the material. Near end effects, in both the cement and Berea, were found to increase compressive strengths, especially when the L/D of the samples was increased slightly over a 2:1 L/D.

Jan 1, 2015

By L. I. Kiss, S. Perron, S. Poncsák, A. Perron

Aluminium industry has an increasing interest in the diminution of the energy consumption of the Hall-Heroult process by the reduction of the anode bottom covering with gas. Electrically isolating carbon-dioxide bubbles, produced at the anode, increase the ohmic resistance of the cell. As the gas production rate is determined by the applied current density, anode covering can be decreased only by a faster evacuation of the gas from the inter-electrode space. Among others, the inclination and curvature of the anode bottom promotes the release of the bubbles and thus decreases the mean value of the covering. Indeed, anode, insert in the molten cryolite bath has a square shape with horizontal bottom. However, inhomogeneous anode consumption, electro-erosion, bath circulation and the inclination of the bath-metal interface modify this shape gradually during electrolysis. Anode bottoms become inclined with rounded comers. This concentrates the gas generation to the center zone of the anode. As the bubble layer cannot be observed directly because of the hostile environment and the opacity of the bath, real scale air-water systems are used to study the impact of the anode geometry on the anode covering and the bubble induced flow. However, the physical properties of such a system are different than that we can found in the real cells. A simulator using the Lagrangian description of the bubble layer has been built and presented in earlier papers. This tool was used to investigate the effect of the anode inclination on gas covering and bath velocity, for both C02-cryolite and air-water system. Results obtained for the second system were compared to experimental data .

Jan 1, 2006

By M. Hajian Heidary

In this investigation, effect of pressure depth on the globulizations of AZ91 magnesium alloy was studied. Cylindrical samples of AZ91 alloy with 1 cm and 2cm in height and 2.5cm in diameter (H/D = 0.4 and 0.8) were prepared, then the samples were hot pressed at 10 and 30% strain at a constant temperature of 200°C; subsequently reheated at 555°C in different times. Results obtained from optical microscopy (OM) showed that the thickness of the samples and consequently, depth of strain affected regions have remarkable effect on the degree of globulization. In addition, it was found that at a constant strain and holding time, structure of thinner samples is more globular than thicker ones.

Jan 1, 2007

By S. E. Malouf, W. Thorpe

"The Chibougamau Explorers property comprises 35 claims or 1,726 acres in La Dauversiere and Rohault Townships along the west shore of Norhart Lake. The property is serviced by a twelve-mile all-weather high-way leaving the main St. Felicien-Chibougamau highway at mileage 121. This road junction will be a railway siding on the C.N.R. St. Felicien-Chibougamau branch line in early 1957. The property lies 30 miles due south from the town of Chibougamau. The mine-sponsored Chibougamau Explorers town has a population of 250 with 28 residences and a well equipped recreation centre.The original showing was found by Rider Storn, a prospector financed by Val d'Or interests during the 1950 field season. This showing did not stand up to detailed prospecting but it did lead to the discovery of the main zone by A. Oakley during the following winter.The company was incorporated in I 950 under the presidency of E. Ayrhart. Financing following the initial drilling involved an increase of capitalization from 3 to 5 million shares and the advent of financing and management control by Anacon Lead Mines Limited.Exploration and development to Jan. I, 1956, has included 42,331 feet of surface diamond drilling; 31,363 feet of underground drilling; a 3-compartment shaft to 1,245 feet, with 8 levels at 150-foot intervals; and I 1,054 feet of drifts, crosscuts, and raises. Ore reserves to the 900-foot level, proven and partly developed, total 546,725 tons averaging 0·30 ounces gold and 0·93 per cent copper. A mill with capacity of 500 to 750 tons will be installed by January 1956.GeologyThe Chibougamau Explorers property lies along the belt of Keewatin volcanics striking through Bachelor Lake, Opawica Lake, Doda Lake, Coapatina, Norhart Lake, and Rohault Lake, where it is in close proximity to the paragneisses of the Grenville series. The boundary of the paragneisses, a gradational contact, lies within a mile of the Chibougamau Explorers shaft collar."

Jan 1, 1949

By J. Yue

The particle breakage kinetics, product size and size distribution, as well as grinding limit for quartz suspensions in a horizontal stirred mill are investigated. Results of grinding studies on quartz suspensions in a horizontal stirred mill confirm "first-order" particle breakage rates. A comparison of equation fits reveal that the product size distribution from stirred mill is best represented by the Rosin-Rammler equation when compared to the Gaudin-Shuhmann equation, but results fit poorly as particle size below five microns. A method of determining breakage rates versus particle size is demonstrated which shows that the 10% top size fraction is most appropriate for the determination of specific breakage rates. The method is used to demonstrate the relationship between particle breakage rate and particle size. As the feed particle size becomes smaller, the size reduction ratio decreases and approaches a grinding limit which depends on the effective stresses created in the mill.

May 1, 2004

By Robert A. Brown

"The O'Brien mine is in the west-central part of Cadillac township along the Cadillac 'break', which extends for 7,500 feet from west to east across the property.The ground was staked in 1924 by A. Dumont and W. Hermeston, prospecting for M. J. O'Brien, Limited. The first vein discovered is that now known as No. 1. The No. 4 vein was later exposed on the surface; the other veins known to be present were revealed by underground work and diamond drilling carried out both from the surface and underground.The mine is developed from three vertical shafts: No .. 2 extending from the surface to 2,000 feet; No. 3 from the surface to 1,500 feet; and No. 4 from 2,000 to 3,000 feet. There is on the property a mill of about 200 tons a day capacity, the usual shops, and a townsite housing about forty-five families. Gold production to the end of September, 1944, amounted to 295,824 fine oz. from ore having an average grade of 0.505 oz. a ton. In addition, the mill recovered 3,237 tons of arsenic trioxide having a purity of 83 per cent As2O3.GENERAL GEOLOGYThe four main groups of rocks of the Cadillac area- the Cadillac, Blake River, Kewagama or Pontiac, and Malartic- are found within the boundaries of the property. However, as these rocks have been fully described in · various government reports (1,2,3,4,5,6) and all the known orebodies occur within one group- the Blake River-a detailed description of the geology will be dispensed with here. The known ore deposits lie south of the Cadillac 'break' in a group of rocks that includes 'greenstones', greywackes, conglomerate, porphyries, and a rock of gabbroic composition that may be intrusive. Both flows and tuffs can be recognized in the 'greenstones'. The two bands of porphyry that trend easterly across most of the property have been considered by some observers to be dyke rocks and by others to be flows. Certain evidence points to their being dykes that were intruded as sills considerably prior to the emplacement of the gold-bearing veins, but the possibility of flow origin cannot be fully rejected. North of the talc-chlorite-carbonate schist zone forming the 'break' the rocks are conglomerate and greywacke with iron formation, in which there are no known veins of ore grade. Within the schist zone are minor acid and basic intrusives. The general trend of both bedding and schistosity is a few degrees south of east. The dip is generally vertica1 or steep to the south."

Jan 1, 1954

A line of nine self-aspirating W emco cells, part of the Rougher circuit at Division Salvador, Codelco Chile, was retrofitted to allow operation with forced air. Each cell could be operated self-aspirated or with compressed air supplied through a 4-inch line equipped with a mass flow meter. Gas dispersion was characterized for the two modes of operation. Metallurgical performance was compared under self-aspirated conditions and with different gas profiles under forced air conditions. Cell characterization demonstrated that with forced-air the cells could not evenly distribute air across the cell volume. Bubble size increased with gas velocity, as expected, with no difference between self-aspirated and forced-air operation. A systematic increase in bubble size down the bank was found. Cumulative bubble surface area flux down the bank showed no major differences between surveys. However, grade-recovery curves revealed that the highest recoveries were invariably obtained with forced air operation. Plotting concentrate mass recovery vs. enrichment ratio showed that the line performance could be controlled over a wider range when forced air was used.

Jan 1, 2007

By K. Tsuchihashi, A. Kishimoto, T. Uda, Y. Kado, S. Maruyama

We recently proposed a continuous production process for titanium metal with some experimental results. In the process, titanium tetrachloride is reduced by Mg in the presences of Bi to form a Bi-Ti alloy at 900ºC. At this temperature, the obtained alloy is liquid and allows for the continuous removal of product and heat from the reaction vessel. The obtained Bi-Ti alloy was vacuum-distilled at 1000ºC for 24 hr, and 98 at % (92 wt %) titanium, which was easily crashed to powder, was obtained. Bi solvent was selected because of the highest titanium solubility at 900ºC and the moderate vapor pressure among the candidate metals (Bi, Sn, Pb, Zn, and Sb).

Jan 1, 2014

By Samuel Nunoo

Gibraltar Mine is an open pit copper-molybdenum mine in central British Columbia. The stability and potential failure modes of pit walls in the Granite Pit are controlled by the orientations of joints and faults. The structural geology also has an impact on as-mined bench geometries. Geotechnical mapping and surveying were conducted using stereo photographs and photogrammetry software. Detailed digital terrain models were constructed to determine the pit wall geometry and the locations and orientations of larger structural features. The pit wall geometry is compared to conventional survey results and the mapped structures are compared to results from a nearby geotechnical borehole.

Aug 1, 2013

By D. F. Sangster, P. J. Hood, G. A. Gross

"Some results are presented of an integrated geological and geophysical study of Precambrian magnetite-quartz iron-formation in the Kapiko Iron Range north of Nakina, Ontario. Detailed geological and topographical mapping was carried out in conjunction with detailed geophysical investigations such as vertical magnetic field, vertical magnetic gradient, hammer seismic, horizontal-loop electromagnetic and magnetic susceptibility surveys. In addition, oriented drill cores were obtained for magnetic polarization studies.Vertical-force magnetic anomalies within individual rock units showed a strong positive correlation with topographic relief. Moreover, magnetic intensity was found not to be quantitatively proportional to the magnetite content in the rocks, due to the presence of high remanent magnetization of the magnetite. Also, the remanent polarization direction was sub parallel to the plunge of the fold structure and mineral lineation in associated rocks.Vertical magnetic gradient measurements were found to be more accurate in locating buried contacts between magnetite-quartz iron-formation and country rock than the vertical magnetic force data.An in-situ magnetic susceptibility meter was used to measure the magnetite content of iron-formation, and the measurements were subsequently compared with vertical-force magnetic results.The results of the EM survey were inconclusive; some magnetite-rich zones produced anomalies in the out-of-phase component. whereas others did not. The reasons for such inconsistencies are not yet apparent.Longitudinal seismic wave velocities measured parallel to the bedding in magnetite-quartz iron-formation ranged from 10,600 ft/sec. to 14,500 ft/sec. The seismic velocity appears to increase with an increase in deformation of the rocks."

Jan 1, 1966

By M. R. KEYS

"The property is about 120 miles north of Red Lake in the Favourable Lake area, district of Kenora ( Patricia part) , Ontario. The original discovery was staked by K. Murray in December, 1927. The first development work was done by the Favourable Lake Mining and Exploration Company. In 1936 Newmont Mining Corporation acquired control of the property and developed it to production by September, 1939. Production to the end of 1945 had a value of $7,534,656 from 441,457 tons of ore, an average value of $17.04 a ton . This production was mainly from gold and silver, with lead and zinc of subordinate importance.The rocks in the vicinity of the mine trend in a north-south direction. From east to west the sequence is: (1) A dark-coloured, massive rock, presumably a flow, composed mainly of quartz and biotite. ( 2) A buff- coloured, scricitized rock locally containing quartz phenocrysts and black hornblende. This is an alteration product of the dark variety and is later than the folding. It occurs mainly within the dark rock. (3) A series of sediments comprising thin beds of greywacke, grit, quartzite, conglomerate, and iron formation. ( 4) A gabbro sill, now largely altered to amphibolite.With the exception of the buff type, these rocks dip 80°E. The buff rock has a very irregular outline in comparison with the other rocks and appears to dip somewhat more flatly east. Felsite and diorite dykes are also present. These have a general strike of S.60°E. and dip 65°S .W. A post-ore diorite dyke trending N.15°E. occurs about 2,000 feet east of the mine area."

Jan 1, 1954

By H. Clare Horwood

"In the Little Long Lac-Sturgeon River area of Thunder Bay district, Ontario, most of the known ore deposits occur in fractures, shear zones, and zones of brecciation, in folded sedimentary rocks . In two or three instances, however, deposits occur in Keewatin greenstones and in later granitic rocks of probable Algoman age.The principal regional features in the area are the long narrow bands of Keewatin volcanic rocks and Timiskaming sedimentary formations that extend east from lake Nipigon to Long Lac, a distance of about 65 miles. The belts continue to the east for many miles beyond Long Lac and probably for some distance to the west under the waters of lake Nipigon. They are intruded in places, particularly in the north, by large masses of diorite, granodiorite, and granite. Small intrusive masses of diorite and porphyry also occur in many localities throughout the area.The Little Long Lac-Sturgeon River mining area is the youngest camp in Ontario and has been in production only since 1934.The district is readily accessible from the branch line of the Canadian National railway between Longlac station on the main line and Port Arthur, and from the Trans-Canada highway No. 17. Both these means of communication traverse the area in a general east-west direction parallel with the principal structural trend of the rocks of the region.In the Little Long Lac section discoveries were made by W. W. Smith in 1931 (Hard Rock property) , by T. A. Johnson in 1932 ( Bankfield property) , by T. A. Johnson and T. Oklend in 1932 (Little Long Lac property) , by F. MacLeod and A. Cockshutt in 1933 (MacLeod-Cockshutt property) , and by George and Frank Papineau in 1935 (Hutchison Lake property).In the Sturgeon River section discoveries were made by Powers and Silam in 1925 (Northern Empire property), T. A. Johnson in 1931 (Dik-Dik property) , and by Russell and James Cryderman in 1935 and 1936 (Sand River and Leitch properties)."

Jan 1, 1954

By J. W. Coetzee, H. E. Deist

Carbon-in-pulp (CIP) and Carbon-in-leach (CIL) processes are used extensively in the gold mining sector to recover aurocyanide from solution. This paper discusses the modelling of these operations, using various simplifying assumptions, in order to simulate the effect of changing operating conditions on the dynamics of the system. It is assumed that the adsorption and leaching rates are a function of concentration difference between the source phase and recipient phase. The simulation was set up using Matlab's Simulink interface. Plant variables are solved using a medium scale Newtonian search algorithm in Matlab, where the differences between the actual and predicted plant outputs are minimized on a continuous time-line.

Jan 1, 2005

By J. L. Nebreda, Z. Fan, J. B. Patel

"A380 alloy has been widely employed to produce thin-wall components via high pressure die casting (HPDC). During HPDC processing, there is a significant problem of property variation even with the very stable and precision automated casting procedure. The enlarged variation lowers the component performance and affects the steady and uniform property distribution along the whole component. In this study, high shear melt conditioning technology was developed and applied to the melt prior to pouring into the shot sleeve. Experimental results show that with application of intensive melt shearing, the variation of elongation was decreased from 21.8to 13.9%, and yield strength variation was reduced from5.5to 3.6%. The improved property variation is attributed to enhanced nucleation both in shot sleeve and the die-cavity, and to the improved distribution of porosity and secondary phases as a benefit of high shear induced grain refinement. With application of high shear, large number of the Spinel phase (MgAl2O4) particles with double size distribution in average diameter of 80and 300nm were formed. Due to the good orientation relationship and small misfit between alpha Al and MgAl2O4, the particles act as the effective nuclei in shot sleeve and die cavity, respectively.INTRODUCTIONA380 alloy had been widely used for produce automotive components by cold chamber high pressure die casting(HPDC) processing, due to its relative good mechanical properties and cast ability performance, and more important most of A380 ingots were produced by recycled Al alloys. Therefore, to promote utility of A380 will contribute to reducing emission of CO2and to the lightweight of automotive industry as well. When cold chamber HPDC was employed to produce the components, the mechanical properties of A380 alloy always have a distribution range (named variation herein), or in other words the properties were not stable. The reason was ascribed to the typical solidification characteristics occurred in HPDC processing. During HPDC processing, there are three steps of melt movement. The first step is that the melt was poured into shot sleeve at temperature of about 680°C, in which shot sleeve was preheated by few shots of melt filling, or by heater. During step one, the first solidification of melt was happened in shot sleeve and about 20% solid made of coarse a-Al grains was formed(Li, Xiong, & Guo, 2016).Such kind of grains will retain in the final products and deteriorate the properties variation. The second step is that the melt was forced moving forward slowly by plunger at about 0.2–0.3m/s until the liquid melt front to the inner gate of mold. The third step is that the melt moved very fast to complete the filling of die cavity and refilling the solidification shrinkage by fast movement of plunger about 2–3m/s. The second solidification was occurred in die cavity and fine a-Al grains will be formed by high cooling rate at about 500–1000°C/s. From the solidification point of view, the grain size was dominated by nucleation and growth. When HPDC was employed to produce components, the melt quality becomes the priority factor to control the solidification process as the other casting parameters were kept at the same."

Jan 1, 2018

By P. J. Oakley, W. M. Thomas, S. W. Kallee, D. G. Staines

A Friction stir welding (FSW) is now extensively used in aluminium industries for joining and material processing applications. The (FSW) technology has gained increasing interest and importance since its invention at TWI almost 15 years ago. The basic principle, recent applications and the continuing development of the FSW technology are described. The paper will introduce some of the variants of FSW, such as Dual-rotation, Twin-stir(TM), and the Stir-lock(TM) technique.

Jan 1, 2006

By R. Chen, K. Xia

In deep mining activities, blasting induced compressive stress wave would be reflected as tensile wave at a free surface around the openings, which would facilitate the tensile failure of the deep rock since it is much weaker in tension than in compression. It is thus critical to understand the dynamic tensile failure of rocks subjected to static tension for deep rock engineering applications. A modified split Hopkinson pressure bar (SHPB) system is adopted to load Brazilian disc (BD) samples statically, and then exert dynamic load to the sample through stress wave generated by impact. The pulse shaper technique is used to generate a slowly rising stress wave and thus to facilitate the stress equilibrium of samples during each test. Five groups of Laurentian granite (with tensile strength of 12.8 MPa) BD samples under the pretension stress of 0 MPa, 2 MPa, 4 MPa, 8 MPa, and 10 MPa are tested under different loading rates. The result shows that the dynamic tensile strength of the rock decreases with the increase of pre-tension stress, which is easy to understand because pre-tension stress weakens the rock sample. It is also observed that under the same pre-tension stress, the dynamic tensile strength increases with the loading rate, revealing the so-called rate dependency that is common for engineering materials. Furthermore, it is found that the loading rate sensitivity of the rock increases with the increase of pre-tension stress. A qualitative explanation for this phenomenon is that more microcracks are generated and activated under higher pretension stress, resulting in a more viscous material. Based on the results, an empirical equation is proposed to fit the data and explain the dynamic tensile behaviour of rock under pre-tension stress. A physical model is used to simulate the dependency of the dynamic tensile strength of rock on the pre-tension stress and the loading rate, which is shown to be in good agreement with the experimental results.

Jan 1, 2015

By M. Pötsch, W. Schubert

"The kinematic analysis of rock blocks is one part of a block stability assessment. It provides information about a block’s movability, i.e. if a block can detach from the adjacent rock mass and move towards the free space. This paper introduces an analytical method for the analysis of pure rotations of arbitrary finite polyhedral blocks. It enables the engineer to judge a block’s rotatability about a corner or an edge. The method is presented using a graphical and computational method. The application of the method is illustrated with an example. INTRODUCTIONThe kinematic analysis of rock blocks deals with the determination of the movability of blocks. Established methods typically cover: (a) translational motion (John 1968, Markland 1972, Londe et al. 1969, 1970, Goodman & Shi 1985, Warburton 1981) and (b) the specific rotational cases Toppling (Goodman & Bray 1977) and Slumping (Kieffer 1998). Mauldon & Goodman (1990, 1996) and Tonon (1998) introduced rotational kinematic analyses for three-joint pyramids and tetrahedral blocks extending Goodman & Shi’s block theory.All mentioned methods are subject to assumptions that limit the applicability to engineering problems. For example, the kinematic test of daylighting joints or intersections inherently assumes a finite block, which only can be proved by the finiteness theorem (Goodman & Shi 1985). Daylighting of an intersection or plane is not a sufficient condition for removability. Goodman & Shi’s and Warburton’s methods are limited to translational block motion. The methods for block rotations so far address tetrahedral blocks only. They also address only pure rotational modes and neglect the general motion of a rigid body. This paper introduces an analytical method for the analysis of the kinematic rotatability of finite rock blocks. It extends Mauldon's and Tonon's theory of rotatability of tetrahedra to arbitrary polyhedral blocks. A graphical method using the stereographic projection and a computational method are presented."

Jan 1, 2015

By Jakob Janzen

This paper demonstrates that the introduction of a Battery Energy Storage System (BESS) into an isolated diesel generator based grid has the potential to reduce the electricity cost and deliver a good return on investment. Energy storage also allows for increased renewable energy penetration, into the isolated grid, that further reduces the electricity cost. The isolated power system studied includes several diesel generators of different sizes. The typical operating pattern is for the generators to be run at 80% to 85% of their power rating to deal with load fluctuations. Operating the diesel generators near their full power rating, however, improves their efficiency, in this case, while the frequency and load control can be carried out by the BESS. Operating the diesel generators near their full power rating reduces the number of generators which need to be operating to support a given load. As a result, the improved efficiency of the diesel generators and reduced maintenance cost, for operating fewer machines, are the reason for the savings.

Aug 1, 2013

By P. Molera

"In Catalonia, between the 10th century and 1878, some installations were developed which at first were called ""factories"" and later ""fargas"", since they manufactured iron and steel of low carbon content. Because of its excellent quality, the fame of this metal spread from Catalonia to many countries. Fargas were started in Corsica, the Low Countries, Germany, Oriental India, Madagascar, Finland and along the rivers Orinoco and Mississippi among others. The extractive process used in these fargas has passed into history as the ""Catalan Farga Process"", ""Catalan Process"", or just ""Farga Catalana"". With this technology, raw material was obtained for the fabrication of tools, nails, weapons, grilles, etc. to supply the national and export markets. Aside from meteorites and native iron, found as metallic nodules in basalts in Greenland and Palestine, the ancient civilizations, which had no means to reach the fusion temperature of iron (1535°C), obtained iron by a direct process. Similar processes were used until quite recently (1930s) by certain tribes in Central Africa; holes in the ground were filled with iron ore and coal, and combustion encouraged by blowing air by mouth or with bellows made of animal skins. Thus without melting the iron it is possible to reduce the mineral and obtain a porous and impure metallic mass, which is purified on forging by mechanical squeezing. With a process very likely similar, the famous column at Dehli weighing 6.4 m.t. was built about 300 B.C."

Jan 1, 1999

PROFIL D'ENTREPRISE Léandre Gervais & Associé(e)s est une société d'experts-conseils en ingénierie de la structure, de la mécanique & de la tuyauterie, et de l'électricité et de l'instrumentation. Nous offrons des services hautement spécialisés à notre clientèle des secteurs industriels des mines et de la métallurgie, au Québec, au Canada ainsi que dans plusieurs autres pays du monde. Notre mission est d'assurer à nos clients un ensemble de services d'ingénierie de haute qualité contribuant à la mise en oeuvre et à la réalisation de leurs projets d'affaires. Léandre Gervais & Associé(e)s emploie 36 employés à temps pleins et opère à partir de Val-d'Or, province de Québec. SERVICES OFFERTS Nous avons réalisé de multiples mandats pour notre clientèle de différent pays du monde dans les domaines d'expertise les plus variés : ?Études de faisabilité ; ?Ingénierie conceptuelle ; ?Ingénierie de détail ; ?Gestion de projets ; ?Ré-ingénierie de la maintenance.

May 1, 2001