Search Documents

By K. H. Concannon, K. I. Hindle

"The use of simulation modeling in the primary metals industry (including mining, smelting, and metals processing) is expanding as significant cost-benefits can be realized by companies employing capital intensive processes. The economic scale and complexity of operations found in this industry is well-suited to applications using this ""what-if' computer modeling technique, ranging from proof-of-concept projects to testing facility design or configuration, and determining resource requirements. More recently this technique is being used to support on-going planning and scheduling for operations using the simulation as the engine to derive the best schedule. With increased computer processing capability, deterministic models are now being developed to capture the workings of a production system or conversion process, allowing real-time monitoring and scheduling of these operations. This paper presents a sample of some recent discrete-event simulation modeling applications that demonstrate the capability and benefits of this technique in the primary metals industry. These include the use of simulation to support the long-range planning of underground mining operations; the modeling of a metals supply chain network that employs a multi-modal distribution system; the scheduling of a metals coil finishing process using a deterministic simulation model; and a more detailed description of the use of simulation to model furnace and converter aisle activities within a smelter operation in Northern Canada."

Jan 1, 2007

By Greg Yuriy

Over the past two decades, Vale Canada has used discrete-event simulation in a key role to support many underground mining studies. Traditionally, general purpose simulation software was developed for manufacturing and service applications. Not until recently have Mine Designers and Simulation Specialists been able to model the dynamics of underground mining scenarios without extensive manipulation of the inherent tools and entities contained within these simulation environments. This paper discusses Vale Canada?s past experiences with a variety of simulation software, describing model development methods and highlighting some of the lessons learned. The paper concludes with a review of recently considered software and provides a framework of desired attributes/functions for modeling mining scenarios.

Aug 1, 2013

By Kristine L. Pankow, Keith D. Koper, Michael K. McCarter, Jared R. Stein

"The Wasatch Plateau region of central Utah contains a number of active underground coal mines in an area of well-documented natural seismicity. As a result, there are several hundred seismic events recorded in this region every year that may or may not be attributed directly to mining activity. Previous studies have linked the majority of this seismicity to mining activity through various means, but have not attempted to develop a method for categorizing individual events within a seismicity catalog. This study uses both relative relocation and waveform cross-correlation to separate events based on hypocentral depth and waveform similarity, respectively, in the area surrounding a single longwall coal mine within the Wasatch Plateau. We conclude that relative relocation can help with depth discrimination, but there are some unavoidable limitations caused by sparse seismic network coverage. The cross-correlation technique is more promising, with the ability to group events by waveform similarity and track mining progress over both space and time. With cross-correlation based clustering, we confirm that the vast majority (97.6%) of events in the study area are directly related to mining activity, but that a small population of natural seismicity is present as well. Eliminating this natural seismicity from the catalog of seismic events, leaving only those events related to mining, can give operators valuable information about the changing conditions as mining progresses and eliminate concern over events unrelated to mining activity.INTRODUCTIONA number of underground coal mines are located in the Wasatch Plateau region of central Utah, and have been in operation since the late 1800s. This activity has resulted in a well-known history of mining induced seismicity (MIS) (Wong, 1985). Additionally, this area is situated near the border of the Basin and Range and Colorado Plateau physiographic provinces, where naturally occurring tectonic earthquakes also occur (Arabasz et al, 1980). Seismicity separating these physiographic provinces is part of a larger band of seismic activity known as the Intermountain Seismic Belt (ISB) that runs from Montana to Arizona and cuts through the western boundary of the central Utah coalfields (Smith and Arabasz, 1991) (Figure 1)."

Jan 1, 2015

By Guangyao Si, Ismet Canbulat, Xun Luo, Yi Duan

Routine analyses of seismicity and event locations are essential for coal burst risk management, especially in the presence of unfavorable and complex geology. In an Australian longwall coal mine, a small-scale seismic monitoring network was installed to monitor a section of a major dyke intersected by the roadway, which was classified as a coal burst prone zone. The Deep Convolutional Neural Network (DCNN) technique was adopted for automatically discriminating seismic events to assess seismicity in this zone. The discriminated seismic events were located with the Shortest Path Method based on the Boundary Discretization Scheme (BDS-SPM) for temporal-spatial seismic event distribution analysis. The results demonstrate that the seismic events are discriminated with an average accuracy of 94%. Moreover, the seismic events are more accurately located using the BDS-SPM with a 3D heterogeneous velocity model compared with results obtained with a homogeneous velocity model. The temporal and spatial variances in seismic activities identify that the rock fractures in the dyke zone are closely related to the longwall mining process and a local dyke has a controlling influence on the stress concentration in its vicinity.

Apr 26, 2022

By Sherwin F. Kelly

THE present lull in engineering activities presents an advantageous moment for inquiring into the position now occupied by geophysics in its various fields of application. The recent over-expansion in all lines of work resulted in an uncritical employment of these methods of exploration which was bound to have its harmful reaction. The pendulum has now swung back, producing a decidedly hypercritical attitude toward the new science that had been recently welcomed so joyfully as the panacea for prospecting troubles. The severe retrenchment in exploration has forced upon us a period of waiting and, it is to be hoped, of profitable contemplation. This can well be employed in trying to pass a reasoned judgment on what position geophysics should occupy in the new geological exploration programs that will inevitably ac- company the looked-for industrial recovery.

Jan 1, 1931

By R. T. Rudd

DISCUSSION R. T. RUDD I have the following comments on Mr Bosman's paper. Before giving the company permission to discharge neutralized mine drainage containing 1800 mg/l of dissolved solids into a public stream, my Department had to consider the matter from both the legal and the practical viewpoint.

Jan 4, 1974

The Role of the Drilling Contractor BY FERRIS E. SAINSBURY Ron Haxby, Occidental Minerals What is the best method for drilling a 12-in. diam hole to a depth of 1200 ft and maintaining the deflection within 25ft from vertical? Ferris Sainsbury: In order to drill a 12-111. hole down to 1200 ft, In most places you would have to employ a rotary tricone bit With a down-the-hole hammer, I expect by that depth significant Inflows of water would be experienced Even with foams, we would not be able to get the water out of a hole of that diameter So the down-the-hole tricone method of drilling would be best. To keep it straight, we would probably employ the largest square drill stem compatible with the diameter of the hole. Ron Haxby: Is it common in drilling contracts to put in a vertical hole clause? Ferris Sainsbury: You can put in the vertical hole clause Ron Haxby: Does it cost more? Ferris Sainsbury: Normally, if the hole deviates, the driller can correct it with deflection wedges and will charge it back to you at cost, plus 10% or somewhere along that range Any drill hole has a tendency to go perpendicular to the dip, and there are times when stabilization systems are not successful in preventing the hole from turning into the dip of the formation. The only solution, in this situation, is to go in with an oriented wedge and kick the hole back out.

Jan 1, 1979

MR. HARRINGTON: I would like to ask Mr. Bottomley if they fire their shots while the men are in the mine? MR. BOTTOMLEY: No sir; the shot-firing law in our State requires the shots to be fired after the men are out of the mine. We fire more shots than we did under the hand method of loading. Where formerly four holes were drilled, now we drill eight, using the same amount of powder as we did formerly, perhaps a little more, but we use a slower grade of powder. CHAIRMAN LITTLEJOHN: Any further discussion, gentlemen? MR. GIBSON: Mr. Chairman, I would like to ask Mr. Bottomley, if, in his shot-firing system, as I understand it, he has a very high coal and is working something like eleven feet? MR. BOTTOMLEY: Something like that. MR. GIBSON: I read the article that you referred to. Did you use both the first and the second delay explo¬ders in shooting the coal? MR. BOTTOMLEY: We use the ordinary fuse. MR. GIBSON: That is what I wanted to find out; whether you use the delay exploders and the detonators. In Utah, we are getting away from the fuse altogether, because we do not think it is accomplishing anything to use the electric light on the cap, and' at the same time shooting with fuse. We are defeating the object we might gain in using the electric safety lamp. That is a point that might be discussed a little further. MR. E. H. JOHNSON, The Coloder Company: I would like-since it is a general opinion that mechanical loading breaks up the coal, to ask Mr. Bottomley, if he finds a larger percentage of small coal, or how much, through the use of the mechanical loaders? MR. BOTTOMLEY: We do not find any great difference. In fact, we have succeeded in improving our lump coal percentage, and that is important, because it is largely domestic. At first, when the machines were installed, the boys shot the coal too hard, and we fell down a little, about two or three per cent, the first month, in our lump percentage, and we tightened up a little and changed our methods in shooting and got on to the stunt of drilling more holes and getting a slower grade of powder. We could not load as much coal in one day as we could before, but we did succeed in in. creasing our lump coal percentage four and a half per cent over the hand method, and that is quite a factor-quite an important factor and since our domestic trade is over with, we loosened up on the shooting and we are running about the same as we did with the hand loading, maybe a half per cent more, or one per cent, but pretty much the same. MR. SHUBART: Your coal is very high? MR. BOTTOMLEY: Yes. MR. SHUBART: Have you tried sunbbing the coal before shooting? MR. BOTTOMLEY: No, we have not tried that. It is very hard coal to shoot; we have no defined cleavage in our coal. It is the worst coal I ever saw to shoot. We do not try to snub it. Formerly, as I said, the miners drilled four holes, pretty well up near the top, and then four holes down, within about three feet of the bottom. It is the same thing as a snub shot, and that we find is the best method for shooting the coal. We have not tried any mechanical device for snubbing as yet. SECRETARY SHUBART: At one mine I saw snubbing tried with a Jeffrey undercutter with a seven-foot Jeffrey bar. This mine, by the way, is very similar to Mr. Bottomley's condition. They have very high coal, so the snub cut is a small percentage. They cut a seven foot undercut and raise the machine on an eight inch tie and undercut five feet to give the effect of a snub cut, 14 inches at the front of the cut, and then six inches at the back of the cut. They made a very marked increase in lump. (The bug dust would be a small percentage of the total coal), and they increased the percentage of the slack at the same time. They found a decrease by reason of the easier shooting of about 15 per cent in their slack. In other words their nut and lump coal ran considerably higher and their slack decreased 5 per cent with the snubbing cut. Mr. Bottomley, I believe, understands those conditions.

Jan 1, 1926

By Jay F. Smith

The Alan Wood BOF Shop consists of two 140 ton furnaces with a rated yearly capacity of 1-1/4 million ingot tons, he hot metal for the BOF Shop is supplied by two 18 foot blast furnaces which produce an average of 2000 tons per day, To achieve the rated ingot ton capacity with this limited supply of hot metal, the preheating of scrap is necessary. This is accomplished in the furnace by using an oxygen-natural gas burner. Piping Arrangement The natural gas and oxygen are piped into the Shop through 8-inch lines to their respective valving stations. As can be seen from Figure 1, the flow of natural gas passes through a flow control valve which is common to both furnaces, after which it divides into two 6-inch lines forming two identical and independent gas control lines, one for each furnace.

Jan 1, 1972

By Wallace W. Wilson

This DaoerAA calls to attention for the first time, to the writer's knowledge, a purported recovery of oil by substantially horizontal gas cycling that is considerably in excess of 50 per cent of the oil originally in place. The average angle of dip along the path of travel of the gas is less than one degree of arc. The best recovery by gas cycling previously brought to the writer's attention1 occurred in the case of the Mile Six Field in Peru, where about 68 per cent of the original oil in place has so far been recovered. In this case, however. the recovery was greatly aided 1)y the high dip angle of about 17.5". An approximate estimate of recovery was made by the method developed by the writer' for the case of the Cook Ranch Field, using an average relative permeability ratio function dictated by past experience with a number of good quality sandstones. A result of about 40 per cent recovery of the original oil in place wa* obtained after the 8.5 pore volumes of cumulative gas cycling undergone by the Cook Ranch Field. Tile writer desires to emphasize what appears to him as the sound engineering practiced at Cook Ranch, and also the care attending the taking of data and the high quality of the calculations made. It is impossible to take issue on factual grounds with any of the conclusions drawn in the paper. It is, however, desired to point out the difficulty of explaining the unexpectedly favorable oil recovery on ordinary theoretical grounds. It would be of interest to know whether relative permeability ratios have been measured on the Cook sand. The operator is inclined to attribute some of the excess benefit to settling of the oil across the bedding planes into the lower part of the sand. It is difficult for the writer to appreciate the importance of such an effect, inasmuch as the concomitant opening of excessive vacated pore space in the upper part of the sand would he expected to cause excessive bypassing of the gas. REFERENCE 1. Welge. H. J.: "A Simplified Method for Computing Oil Recovery by (Gas or Water Drive." Trans. AIME, (1952), 195, 91. AUTHOR'S REPLY TO MR. WELGE Welge has poined out clearly the difficulty of applying material-balance procedures in calculating the Performance of low-pressure, dispersed gas injection operations. There can be no argument that even with the most favorable assumptions of K,/K,, and with cumulative gas cycling in excess of normal expectations, the calculated oil recovery will be considerably less than the 72.5 per cent recovery realized from this project to Jan. 1, 1951. There probably are at least two reasons for the unusually high recovery from this reservoir, the effects of which would be difficult or impossible to calculate by known methods: 1. The operator did not use the same injection wells for the entire operation, and exercised considerable control over producing GOR by changing the locations of injection wells whenever well performance indicated excessive bypassing. Fifty-six wells were used for injection, not over 29 of which were in use simultaneously. This procedure resulted in frequent shifting of the fluid movement patterns in the reservoir and undoubtedly increased the ultimate recovery. The writer knows of no satisfactory procedure for calculating the effect of this practice. 2. Vertical movement of oil across the bedding planes of the reservoir rock hy gravity drainage tended to maintain a relatively high oil saturation, and hence a favorable K,/K, ratio, in the lower sections of the reservoir. As pointed out by Welge, this condition has aggravated the tendency for high producing GOR, but at the same time it has resulted in increased oil recovery. As may be seen from the performance charts for the various battery areas, operating (GOR'S have been quite high for the last several years of reported history.

Jan 1, 1952

By E. Tulcanaza

M.S. Azun I do not at all agree with the basic points of the author's conclusion. The use of lognormal or normal model to respond to the attribute distribution function should be carefully questioned. If fitting a distribution function of regionalized variables is the main purpose, one may employ one of certain well-known families of distributions, such as Tukey's lambda functions. For instance, the distribution function of gold assay values studied by Krige (Krige, 1978) may fit the lognormal model well, but one should not exclude that the basic idea of using the distribution function of observed data is to have insight of the physical mechanism that gives rise to the regionalized variables. In Applied Geostatistics, there are many examples considering only lognormal or normal model for the distribution function of geostatistical data even if the observed distribution function has more than one mode. Of course, using lognormal or normal model always brings easy computation for the statistical properties of attribute under investigation. Still, I object to adopting any of those models without having any "statistical inference." As seen from the author's paper, lognormal and normal model produce the same average grade (0.746% copper) and almost the same sample variance. Therefore, which model explains the probabilistic behavior of ore deposit is always an important question that should be replied by the geostatistician using not the common sense but the statistical and probabilistic methods. For kriging procedures, such as linear kriging, one should describe either second order stationary properties or bivariate properties of regionalized variables. Second order properties, such as correlogram function or semivariogram function, are the inputs of the so-called kriging equations. Selecting one of any useable models for the observed second order properties, such as spherical model, is not easy since the models are not based on the correlations structure of the regionalized variables. How those models being used in Applied Geostatistics can be distinguished is another important problem. One may develop many fitting models to respond to the observed second order properties of regionalized variables. However, I suggest that the model should be based on the probabilistic behavior of geologic process. The author used spherical model for copper, wolfram, and silver grades from isotropic sector (Fig. 2, 4, 3b) and exponential model for silver grades obtained along raises in isotropic sector (Fig. 3a). It seems that those semivariograms given in Fig. 3b for isotropic sector and Fig. 4 may be described by the random model implying that regionalized variables do belong to renewal process (Azun, 1983). If I knew the total number of data points used in estimation, I may prove that my conclusion is correct. Recalling the test for the first order sample correlogram value, the test procedure is introduced such that the independence is rejected if [ ] where Za/2 is the a/2 percentile of standard normal random variable and N is the total number of samples used in estimation (Azun, 1983). For the other observed semivariogram functions I suggest the author might try to use the so-called "Markovian model" describing not only the correlation structure of regionalized variables, but also the physical mechanism that produces the regionalized variables. The Markovian model for correlogram function is, [ ] The Markovian model for the other second order properties of ReV are also derived. T ve B, in the above equation, show, for example, structural change and mineralizational variation in a considered deposit, respectively (Azun, 1983). In selecting any model, it is not easy to search for the "best" response to the observed second order properties of ReV's. The Markovian model, based on a theoretical understanding of underlying mechanism, gives more information about the occurrence of regionalized variables and respond to all properties. Random model and the so-called hole effect structure can be easily defined as a special form of the Markovian model (Azun, 1983). An estimation of any second order properties of regionalized variables may be computed through (N-1) lag. However, the dependency between the regionalized variables may be deemed to have "died out" after the so-called stationarity width (range). In practice, the estimation is carried out through 25% of the total number samples available. Therefore, the large fluctuations around the zero level and variance for correlogram or covariogram function and semi-variogram function, respectively, cannot be observed. Since the number of pairs involved in the estimation of higher second order properties is small, the estimation variance at those lags is large. There is no need to show higher order estimated second order properties. After using one of any kriging procedure for block estimation, the distribution function of average grades has the same mean as the drilling sample grades but

Jan 3, 1985

By M. S. Azun

I do not at all agree with the basic points of the author's conclusion. The use of lognormal or normal model to respond to the attribute distribution function should be carefully questioned. If fitting a distribution function of regionalized variables is the main purpose, one may employ one of certain well-known families of distributions, such as Tukey's lambda functions. For instance, the distribution function of gold assay values studied by Krige (Krige, 1978) may fit the lognormal model well, but one should not exclude that the basic idea of using the distribution function of observed data is to have insight of the physical mechanism that gives rise to the regionalized variables. In Applied Geostatistics, there are many examples considering only lognormal or normal model for the distribution function of geostatistical data even if the observed distribution function has more than one mode. Of course, using lognormal or normal model always brings easy computation for the statistical properties of attribute under investigation. Still, I object to adopting any of those models without having any "statistical inference." As seen from the author's paper, lognormal and normal model produce the same average grade (0.746% copper) and almost the same sample variance. Therefore, which model explains the probabilistic behavior of ore deposit is always an important question that should be replied by the geostatistician using not the common sense but the statistical and probabilistic methods. For kriging procedures, such as linear kriging, one should describe either second order stationary properties or bivariate properties of regionalized variables. Second order properties, such as correlogram function or semivariogram function, are the inputs of the so-called kriging equations. Selecting one of any useable models for the observed second order properties, such as spherical model, is not easy since the models are not based on the correlations structure of the regionalized variables. How those models being used in Applied Geostatistics can be distinguished is another important problem. One may develop many fitting models to respond to the observed second order properties of regionalized variables. However, I suggest that the model should be based on the probabilistic behavior of geologic process. The author used spherical model for copper, wolfram, and silver grades from isotropic sector (Fig. 2, 4, 3b) and exponential model for silver grades obtained along raises in isotropic sector (Fig. 3a). It seems that those semivariograms given in Fig. 3b for isotropic sector and Fig. 4 may be described by the random model implying that regionalized variables do belong to renewal process (Azun, 1983). If I knew the total number of data points used in estimation, may prove that my conclusion is correct. Recalling the test for the first order sample correlogram value, the test procedure is introduced such that the independence is rejected if [Ir(1) I > Za/2 N1/2] where Za/2 is the a/2 percentile of standard normal random variable and N is the total number of samples used in estimation (Azun, 1983). For the other observed semivariogram functions I suggest the author might try to use the so-called "Markovian model" describing not only the correlation structure of regionalized variables, but also the physical mechanism that produces the regionalized variables. The Markovian model for correlogram function is, [p (h) = T 1ih , h>1 ,] The Markovian model for the other second order properties of ReV are also derived. [T veß], in the above equation, show, for example, structural change and mineralizational variation in a considered deposit, respectively (Azun, 1983). In selecting any model, it is not easy to search for the "best" response to the observed second order properties of ReV's. The Markovian model, based on a theoretical understanding of underlying mechanism, gives more information about the occurrence of regionalized variables and respond to all properties. Random model and the so-called hole effect structure can be easily defined as a special form of the Markovian model (Azun, 1983). An estimation of any second order properties of regionalized variables may be computed through (N-1) lag. However, the dependency between the regionalized variables may be deemed to have "died out" after the so-called stationarity width (range). In practice, the estimation is carried out through 25% of the total number samples available. Therefore, the large fluctuations around the zero level and variance for correlogram or covariogram function and semivariogram function, respectively, cannot be observed. Since the number of pairs involved in the estimation of higher second order properties is small, the estimation variance at those lags is large. There is no need to show higher order estimated second order properties. After using one of any kriging procedure for block estimation, the distribution function of average grades has the same mean as the drilling sample grades but

Jan 1, 1986

By L. O. Howard

R. HATCH, Garfield, Utah.-The object of the meeting, I believe, is to determine the proper chemical composition to insure long wear of a ball in a mill, This might also apply to rods. I feel that there is only one way of determining accurately the wearing qualities of balls in mills, and that is by actually wearing out 400 or 500 tons of different kinds of balls. Two things add to the wearing qualities of a ball: (1) the tenacity, with which the particular particles of iron or steel are cemented together; (2) the individual hardness of each particle. My experience is that a ball that has a Brinell of 550 to 650 will give comparatively long life. There is considerable difference of opinion as to the best steel and iron alloys necessary to produce a satisfactory grinding ball. Some favor chrome, some manganese and some nickel. Personally, I believe cast iron alloyed with chrome makes the best ball for small sizes; for large sizes (3 1/2 in. and up) steel alloyed with manganese is best. L. 0. HOWARD, Pullman, Wash.-How would this method apply to lining? A. KissocK, New York, N. Y.-I do not subscribe to that exactly. As I am connected with the Climax Molybdenum Co. you would hardly expect me to. We have been much interested in trying to work out this grinding-ball problem and have spent a great deal of thought and effort on it. As we are associated with the American Metal Co. and have a number of properties of our own, we are in position to try our ideas on ourselves first. We have tried different compositions, but have not done anything very remarkable with them yet, except that we have been able to say that the combination of chrome and molybdenum has given the best wearing ball. Mr. Hatch feels that iron is the best medium for grinding, but since toughness as well as hardness is a factor in wearability we feel that a well made steel is better for this purpose. Another difficulty with cast, or chilled, iron is lack of uniformity from day to day or from heat to heat and the impossibility of producing balls free from blow or shrinkage holes which materially affect their life. If cast iron is made hard it becomes brittle and lacks stretch or toughness. This prevents its use in the larger sizes, as breakage is a serious handicap to efficiency. We have tried hard and brittle balls and softer but very tough balls and have proved that maximum hardness, with sufficient toughness to prevent breakage, is the most important factor in resistance of a steel to abrasion. It is well known that chromium is one of the best hardening agencies for either iron or steel, and molybdenum is recognized as an element capable of introducing toughness. In other words, the proper combination of chromium and molybdenum permits unusual hardness with still sufficient toughness to stand up under severest impact. It is yet too early to give any actual mill results of this new combination and though we are proving our ideas at our own properties we have to depend on others to help us make the data more general and more worth while and would like to get some tests

Jan 1, 1928

By G. Johnston, P. K. Chatterjee, S. Holguin

M. Gershon The article is an excellent one, providing a basis for the conduct of the major tasks in underground mine planning and scheduling. I object to only one word. Since that word is featured in the title, it is a strong objection. The word used erroneously is optimization. There is no optimization performed in the article. By definition, an optimization is an approach that yields the best of all possible solutions. We can limit this definition to its narrowest sense, the optimization techniques of operations research such as linear programming, or broaden it to include less sophisticated search procedures that find the best mine plan or schedule. No such claim is made in the article. No optimization of any kind is performed. The major tool used is a simulation, which, in truth, is probably more appropriate for this problem than any optimization. But simulation does not optimize. It only allows for quick evaluation of many options, leading to the selection of the best plan "of those tried," not "of all possible plans." This article, of course, is not the first mining paper to make this error. It seems that the word optimization is used more to impress people than to refer to any act of optimizing. It is used in a way that implies "try to optimize"or "come close to optimize," but not to really do it. It is usually appropriate to quietly ignore this error. But when Dr. Chatterjee, who has done so much good work in true optimization projects and whose work I have come to admire, makes such an error, it is time to publicly dispute it. Having discussed this very point with him in the past, I am sure that he will agree. If he can support this position in his response, perhaps people will start to listen and we can reduce the number of false claims of optimization. This should be easy for him in this case since the approach he used, simulation, is the more appropriate one anyway.

Jan 8, 1988

By M. Gershon

The article is an excellent one, providing a basis for the conduct of the major tasks in underground mine planning and scheduling. I object to only one word. Since that word is featured in the title, it is a strong objection. The word used erroneously is optimization. There is no optimization performed in the article. By definition, an optimization is an approach that yields the best of all possible solutions. We can limit this definition to its narrowest sense, the optimization techniques of operations research such as linear programming, or broaden it to include less sophisticated search procedures that find the best mine plan or schedule. No such claim is made in the article. No optimization of any kind is performed. The major tool used is a simulation, which, in truth, is probably more appropriate for this problem than any optimization. But simulation does not optimize. It only allows for quick evaluation of many options, leading to the selection of the best plan "of those tried," not "of all possible plans." This article, of course, is not the first mining paper to make this error. It seems that the word optimization is used more to impress people than to refer to any act of optimizing. It is used in a way that implies "try to optimize" or "come close to optimize," but not to really do it. It is usually appropriate to quietly ignore this error. But when Dr. Chatterjee, who has done so much good work in true optimization projects and whose work I have come to admire, makes such an error, it is time to publicly dispute it. Having discussed this very point with him in the past, I am sure that he will agree. If he can support this position in his response, perhaps people will start to listen and we can reduce the number of false claims of optimization. This should be easy for him in this case since the approach he used, simulation, is the more appropriate one anyway.

Jan 1, 1987

By G. Le Bel

Discussion by G. Le Bel The paper proposes three methods to determine optimum lifetime of a mining project 1. The conclusion is that the three methods arrive at the same optimum production rate (by coincidence). It is also stated that: "The stochastic and option pricing models, in addition to providing a more realistic optimum mine life than the traditional NPV (net present value) calculation, showed that the example project has the potential for substantially higher return. By explicitly considering a project's potential risks, the stochastic or option pricing techniques can more accurately describe its probable investment performance." This discussion is intended to demonstrate that: •It is no coincidence that the three methods arrive at the same optimum production rate. • The stochastic and option pricing models do not provide a more realistic optimum mine life than the traditional NPV calculation. •The potential for a substantially higher return is misrepresented. First, a review of investment decision theory in mining is needed. Net present value is the preferred performance measure because of the flexibility offered when comparing various alternatives. However, it presents the following disadvantages: • The discount rate selection is not straightforward. • The NPV obtained represents a deterministic characterization of the variables included in the cash flow calculations, i.e. a single-point estimate. Discount rate selection The opportunity cost of capital represents the rate of return that would be obtained if the capital was instead invested in other opportunities considered to exist. Such opportunities must be bypassed if the available capital is invested in the project studied 2. If these other opportunities carry risk similar to the project under analysis (for example, if they are other mining projects), there is no need to add incremental [ ] risk factors. On the other hand, if these opportunities are represented by the acquisition of government bonds yielding 7%, a risk premium factor must be included. The types of risks associated with a mining venture are shown in Table 1. Together they represent the "business risk" of being a mining company. The individual project risk is important to investors if it differs substantially from the business risk of the company and that of the other investments in mining ventures that we must pass up to invest in this particular project. A risk premium or discount would then be required. This can be illustrated by the following example. Consider that a company has two investment alternatives: one is the Cavender hypothetical project located in Nevada, and the other is the same project located in a civil war zone of central Europe. The second project requires a risk premium factor on the discount rate. Conversely, a project showing highly uncertain returns that are negatively correlated with the returns of the firm's other assets may decrease the company's corporate risk. Deterministic character of NPV calculation The cash flows used in NPV calculations are obtained from a set of variables. These variables are estimated using one parameter obtained from the engineering studies. Each parameter is thus a single value that is often termed a "realistic" or "best estimate" as opposed to "optimistic" or "pessimistic." However, considering that each variable can assume a range of values between the

Jan 1, 1994

By N. D. Policow, J. S. Orphanos

D.A. Dahlstrom In the paper, a comparison was made between flowsheets using the screen bowl centrifuge and disc filters, respectively, as the primary dewatering device. The authors very effectively show that by raising feed temperature, mechanical dewatering is enhanced through reduced viscosity of the water. When a final coal moisture of 10% is required, the net effect of heating the feed is a reduction in thermal requirements because of the lower evaporated load on the dryer. The flowsheet comparison on this, however, was not completely correct. Disc filters do not generally utilize enhanced mechanical dewatering through increased temperature. Horizontal belt filters, through the optimal use of steam heating of the filter cake, have shown significant cost reductions over disc filters where steam is not used. In addition, the authors forgot to mention that there is a certain fraction of the feed that is so fine that it cannot be recovered by the centrifuge economically. That is the function of the filtration device shown in Fig. 4. In subbituminous pipeline coals, this fraction represents more than 10% of the feed. Even after flocculation and belt pressing, the moisture content in the fine fraction is 30%-40. Recent tests on a subbituminous pipeline coal slurry, compared an Eimco-Extractor horizontal belt filter with the screen bowl centrifuge flowsheet shown in Fig. 4. Mechanical dewatering was enhanced through the use of a steam dry zone after the initial dewatering zone. Vacuum pump horsepower was reduced by more than 50% through the use of condensors, and heat recovery was utilized to minimize thermal costs. Thermal drying to 10% moisture by weight was required in both flowsheets. The 0.5 wt% fines in the filtrate were flocculated and recycled to the filter, thus eliminating the belt press and solids mixers. When the three major operating costs (which normally account for 50%-75% of the coal operating costs) of steam, electricity, and flocculants at current prices were calculated, the flowsheets utilizing the Eimco-Extractor had cost savings vs the centrifuge flowsheet of 89¢/t (81o; per st) of surface dry coal. This advantage increases with increasing thermal and electricity prices. I congratulate the authors for the strides made with the screen bowl centrifuge. However, filtration flowsheets continue to be improved and their simplicity and low thermal and electrical requirements should not be so quickly dismissed. ? Reply by N.D. Policow From a technical standpoint, a number of extensive test programs have demonstrated that by raising the feed temperature, mechanical dewatering is enhanced whether it be centrifuges or filters performing the separation. Economic constraints usually become the over-riding issue. If an inexpensive source of thermal energy is available, i.e., waste steam, etc., the approach becomes commercially acceptable. For the specific case of coal slurry pipelines, since the dewatering plants will be located close to power generation stations, the preheating approach becomes especially attractive. The cost comparison shown in Fig. 7, which was prepared by a major West Coast engineering firm, contrasted screen bowl centrifuges and disc filters. In both alternatives, a thermal dryer was specified to produce a contract moisture of the final product. Sizing of the thermal dryer on the screen bowl alternative was substantially smaller than that of the filter, since the screen bowl was credited with a four to six percentage point decrease in product moisture compared to the disc. This comparison was made at ambient temperatures and reflected solely the mechanical dewatering capabilities of the two devices. No preheating of the feed was included in this particular comparison. Early generation screen bowl centrifuges operated under the principle of maximum bulk tonnage throughput. Second and third generation machines have followed the trends dictated by the coal industry calling for higher one-pass recoveries than previously demanded. The current generation of high-speed, long-bowl centrifuges permit recoveries of ultrafine solids directly comparable to commercial filter operation. On typical coal preparation plant feeds with nominal 600 E. an X 0 (28 mesh X 0) clean coal, recovery levels in excess of 99% are now obtained. The material that does escape the centrifuge and reports to the main effluent is typically -10 µm, high ash material. We have documented substantial beneficiation effects in screen bowl centrifuge performance with the cake product containing lower ash levels than the feed. On the referenced pipeline, coal tests performed on Western subbituminous coal, we found that the percentage of -10 µm material in the feed was six to eight times greater than normally encountered on eastern Appalachian seams. Our 45-t/h (50-stph) production centrifuge was never set up to maximize recovery. It was used as part of the two-stage dewatering circuit that had been chosen by designers of the proposed pipeline. Recoveries up to 95% were recorded. With field adjustments, that performance could most certainly have been improved. In contrast, manufacturers of the horizontal belt filter provided a small pilot-scale unit that operated on a side stream at the test facility. Numerous combinations of chemical additions were evaluated to determine those that had the best impact on filter performance. Whereas the screen bowl centrifuges operated without chemical addition, producing a dry, fluffy cake, the belt filter required two stages of chemical addition to attain satisfactory performance and high recovery levels. On the two-stage circuit that incorporated a belt press as the secondary

Jan 12, 1983

By V. Miller

C. Hertzler The difference of means of milling characteristics by rock and ore type is tested in a statistical manner in the paper. The null hypothesis being tested is that the sample means of two different populations are equal. It is assumed that the null hypothesis is true when performing the test. Because of this, acceptance of this hypothesis leads to the conclusion that the sample means of the two different populations are equal (or there is no significant difference between the sample means), not that the samples come from the same population, as the author asserts. Rejection of the null hypothesis leads to the conclusion that the sample means of the two different populations are not equal. A P value for this test is reported in Table 2 and is incorrectly identified as the probability that the two samples came from different populations. Actually, what was recorded was 100(1 - p) where p is the usual p-value. The usual p-value measures, on a scale of zero to one, the extent to which the data agree with the null hypothesis. It is the probability of observing a value of the test statistic (Z score) that is at least as contradictory to the null hypothesis as the one computed from the sample data (McClave and Dietrich II, 1982). The author makes no mention of the fact that were these tests to be conducted together, the chances of determining that the null hypothesis is not true when it is really true are increased considerably (Snedecor and Cochran, 1980). This is a multiple comparisons problem and should be dealt with accordingly. The usual methods involve adjusting the level of the individual tests. Table 2 reports an incorrect 100(1 - p) value in the comparison of concentrate grades for other rock types with .30 to .69 % Cu vs other rock types with .70 + % Cu; 86 % should be 98.6 %. Although the article can be criticized for poor data presentation and discussion of that data, the point the author makes regarding the use of these statistics in the ore reserve estimation process is well taken. It does not appear that inaccuracies in the statistical interpretations have hindered that usefulness. References McClave, J.T., and Dietrich, II, F.H., 1982, "Statistics," Dellen Publishing Co., San Francisco and Santa Clara, CA, 766 p. Snedecor, G.W., and Cochran, W.G., 1980, "Statistical Methods," The Iowa State University Press, Ames, IA, 561 p. Reply by V. Miller After reading Ms. Hertzler's comments, it appears that the differences noted were basically semantics. If the paper is ever presented at a math symposium, some changes will be needed. When deciding the best way to present the data to mining engineers and geologists, I decided that presenting the probability that the two sample groupings came from different populations (reverse to the Null hypothesis) was easier and would require less verbage to explain. Hence, the 100(1 - p) difference which was noted. Ideally, I would have preferred to give the recovery and concentrate grades, then comment which ones had significant differences. Unfortunately, Kennecott Corp. did not grant permission to publish the ore body's actual recoveries and concentrate grades. The conclusion is that there was, as expected, a relationship between the ore's geologic properties and the metallurgical properties, and by determining this relationship, the overall mineral model was greatly improved.

Jan 8, 1984

By M. R. Yalamanchili, J. D. Miller

Discussion by M.R. Yalamanchili and J.D. Miller The authors, M. S. Celik et al., should be recognized for their efforts to describe the flotation behavior of boron minerals. In the case of borax and other soluble salt minerals, analysis of the flotation chemistry has been difficult because of the high ionic strengths associated with these soluble salt systems. However, considerable progress has been made in this area, and recently a surface charge/collector colloid adsorption model was proposed by Miller and his coworkers to explain the collector adsorption phenomena observed in soluble salt flotation systems (Milleret al, 1992; Yalamanchili et al., 1993; Miller and Yalamanchili, 1994; Yalamanchili and Miller, 1994a: Yalamanchili and Miller, 1994b). In this work, the sign of the surface charge of alkali halides in their saturated brines was established on the basis of nonequilibrium electrophoretic mobility measurements by laser-Doppler electrophoresis (Miller et al., 1992). Generally, these results are what would be expected from the simplified lattice-ionhydration theory. This electrokinetic information coupled with the stability and prevalence of collector colloids in such soluble salt flotation systems indicates that the selective flotation of alkali halides is due to the adsorption of oppositely charged collector colloids by heterocoagulation. Experimental flotation/bubble attachment results for 21 different alkali halides (Yalamanchili et al., 1993; Yalamanchili and Miller, I994b) confirmed that the flotation response of soluble salt minerals with weak electrolyte collectors can best be explained by the adsorption of oppositely charged collector colloids rather than by the adsorption collector ions and/or neutral molecular dipoles as originally suggested by many researchers (Fuerstenau and Fuerstenau, 1957; Schubert, 1967; Roman et al., 1968). In addition, the flotation of certain alkali oxyanions (Pizarro et al., 1993) and double salts such as schoenite and kainite can be explained by the same collector colloid adsorption mechanism (Miller and Yalamanchili, 1994). The borax flotation results reported by Celik et al. need to be examined in terms of the above mentioned surface charge/ collector colloid adsorption model. Unfortunately, the authors seem to be unaware of this recent work that nicely describes soluble salt flotation with weak electrolyte type collectors such as amines and carboxylates. In view of our past work, the flotation characteristics of borax were of particular interest, and, in this regard, the results of dodecyl amine flotation of borax reported by Celik et al. have been examined in further detail in the light of experimental results from our laboratory. In our research, a vacuum flotation technique was used to study the flotation response of borax (Na2B407.10H20), which has a solubility of 39 g/L at 25 °C) with dodecyl amine hydrochloride as collector. These chemicals were purchased from Eastman Kodak and used as received. Saturated solutions of borax at desired pH values were prepared by continuously stirring the salt solutions over a period of about 10 hrs. It should be mentioned that the conditioning time to achieve equilibrium is an important variable and can significantly change the flotation response of some soluble salts (Yalamanchili et al., 1993). Collector was added to the saturated borax solutions containing about one gram of 100x 150 mesh borax particles, and conditioning was done for about 20 minutes prior to flotation. The borax flotation recoveries from saturated brine are presented in Fig. 1 as a function of collector addition at the natural pH of 9.3, as reported both by Celik et al. and as measured in our laboratory. In addition, the region of precipitation for the dodecyl amine hydroborate is included in Fig. 1. It can be seen in Fig. 1 that the flotation response curves are separated by about one order of magnitude in R12NH3CI collector addition. The flotation results of Celik et al. show that the maximum borax recoveries can be obtained below the solubility limit of the dodecyl amine hydroborate collector. However, in our experiments borax flotation seems to occur only after the precipitation of the dodecyl amine hydroborate collector as might be expected from the collector colloid adsorption model (Yalamanchili et al., 1993) if borax were negatively charged. Further analysis by nonequilibrium and equilibrium electrophoretic mobility measurements for borax indicates that borax is negatively charged at the natural pH of 9.3, as discussed below. The reliability of the nonequilibrium electrophoretic measurements has been demonstrated previously for alkali halides and alkali oxyanions (Miller et al., 1992; Miller and Yalamanchili, 1994). The equilibrium and nonequlibrium electrophoretic measurements for borax were found to be consistent and are presented in Table 1. These results provide clear evidence that borax carries a negative surface charge in its saturated brine (pH 9.3), and the sign of the surface charge of borax reverses and becomes positive if the pH is reduced to 8.6. The equilibrium between borax and its saturated brine can be described by the following reaction: [2Na2B407.1OH2O-4Na++B407=+HB4O7 +OH+19H20] It appears that the oxyanions of the borax lattice provide

Jan 1, 1995

M.S. Azun I have many objections to the content of the author's paper. Before discussing it, however, I would like to repeat the property of semivariogram function. Second order stationary properties of regionalized variables (ReV's) such as semivariogram function y(h) are perfectly known in geostatistics. Also, the kriging equations in the language of mathematical statistics using second order stationary properties are well understood. However, the way to use the sample (estimated) semivariogram function in any one of the kriging procedures is vague. The sample semivariogram function is given as follows: [ ] where N is the total number of samples, Xi is the sample value at the i - th location, Xi+h is the sample value at the i + h - th location, and h is the distance among the samples. An estimation variance of sample semivariogram function of first lag is smaller than that of higher order lag. The theoretical semivariogram function reaches the variance of samples asymptotically. But this is not easily observable because of the larger variation involved in the estimate of semivariogram function. In general, an estimation procedure is done for h = 0, 1, 2, up to the greatest integer less than N/2, even though sample semivariogram function can be computable through N-1. After estimating semivariogram function, the critical question of how to model sample semivariogram function arises. As seen in the above equation, sample semivariogram function is discrete and can be smoothed by the model being selected. Therefore, modeling of sample semivariogram function is the most important step in geostatistics. It not only smoothes a discrete function but also affects the results of the kriging procedure. When the only aim is to model the semivariogram function, which is the basic point of the author's paper, one can employ any fitting techniques, such as curve fitting, or any arbitrary functions, which are called submodels in the paper. The term "arbitrary function" is used rather than "submodel" because there is no basic understanding of developing them. The author suggests that the sum of those submodels can also be used for the modeling of sample semivariogram function. The combination of any arbitrary functions brings many problems instead of giving an insight of the domain structure considered. The author used two arbitrary functions and the nugget effect in response to sample semivariogram function (Fig. 10). For the same example, he stated that the parameters involved in the mixed arbitrary function model can be accepted when the discrepancy between sample semivariogram function and the model is small visually. For verifying the fitting behavior of any selected model, one should not be contented with the visual satisfactory. Some statistical measure such as goodness of fit has to be used. The author's practice is no more than an exercise in curve fitting without any fundamental understanding or conceptualization of the underlying physical mechanism. Furthermore, the selection of any model is not an easy task if the purpose is the search for the "best" response to the observed second order properties of ReV's. I suggest that the Markovian model (Azun, 1983), on the basis of a theoretical understanding of underlying mechanism, which gives more information about the occurrence of regionalized variables, is used to respond all properties of ReV's. There are a lot of problems for modeling of one-dimensional sample semivariogram function. Thus, it is not appropriate to go to higher order dimensional sample semivariogram function modeling. In the meantime, I would recommend that one can connect the values of standardized sample semivariogram function rather than simple values of semivariogram function in the two-dimensional estimation. The standardized values can be computed in dividing the semivariogram function value by the number of sample pairs involved in each lag regardless of the directions. In conclusion, geostatistics is an interdisciplinary area in mining that uses the principles of mathematical statistics. Thus, it should not violate any probabilistic and statistical rules. When Matheron was developing the theory of geostatistical study in the early years of geostatistics, many mining people had a reservation accepting the geostatistical tools. However, this does not mean that we, the geostatisticians, might try to convince those people using some "strange" tools or rules as some authors implied (Baafi and Kim, 1984). Instead, we have to develop and explain the geostatistical tools staying only in the framework of statistical concepts and properties. References Azun, M.S., 1983, "Stochastic Process Modeling of Spatially Distributed Geostatistical Data," Columbia University, Ph.D. Thesis. Baafi, E.Y., and Kim, Y.C., 1984, "Discussion - Comparison of Different Ore Reserve Estimation Methods Using Conditional Simulation," Mining Engineering, Vol. 36, No. 3, p. 280. Reply by J.M. Rendu The interactive method proposed by Rendu allows practitioners to develop semivariogram models that take into account not only the numerical information obtained by sampling, but also highly significant additional information that often cannot be quantified. The geology of the deposit - including hypotheses concerning its genesis, sampling methods, assaying methods, and mathematical methods used to calculate the semivariograms - all have an influence on the numerical results obtained and on how these results should be interpreted. If all the information concerning the spatial distribution of values in a mineral deposit was contained in the sample values, it could be argued that statistical techniques alone would produce optimum models. However, this is rarely, if ever, the case. Methods that allow the user to take into account his experience and his geologic understanding of the deposit should not be rejected for the sake of theoretical statistical purity.

Jan 11, 1985