Kenneth F Lane CUTOFF GRADES IN THEORY AND PRACTICECHAPTER ONE Introduction When I first joined the mining industry I believed that ores, like buried treasures, possessed immediately recognisable characteris- tics; they glowed in the dark, or glittered in torchlight, or were black and strangely heavy. This is a preconception which is shared by most Jaymen and I was effectively a layman at the time since my training had been in mathematics and economics, not in mining. Even after working at several mines, my conviction was unshaken, When the geologists pointed to rich seams in the hanging walls or chipped what they proclaimed as good ore samples from new headings, I attributed my inability to distinguish the material from any of the surrounding rock to my shameful ignorance of mineralogy. I was impressed; these people were sensitive to subtle differences which no amount of concentration revealed to me. ‘Then, inevitably as a mathematician, 1 became concerned with the statistics of sampling. 1 learned that samples are the nerves of operating mines, borchole samples, chip samples, channel samples, grab samples, dust samples — thousands of samples which are regularly taken, assayed, plotted and interpreted. ‘They are the ‘means of mine control and, in the grade control department, they are the means for determining the limits of the ore. At last I understood. Although some ores may be distinguishable by certain physical properties, ores in general are defined operationally by a cut-off grade; material with a mineral content above the cut-off is scheduled for treatment, other material is left or dumped as waste Having made this discovery, I asked what seemed to me to be the obvious next question:Economic Definition of Ore Why work to this particular value of cut-off grade rather than some other value? ‘Typical answers were: We have always worked to 0-3%. Head office decided 5% combined metals some years ago. ‘That is a technical matter; we leave it to the people on-site, I think several cut-offs were examined in the feasibility study and 1% seemed best. I guess our costs are running at $10 a tonne and uranium is worth $10 a pound, so 1 Ib/tonne must be about right. ‘The inadequacy of these answers stimulated my original interest in the definition of ore. The fact seemed to be that the subject was not clearly within the scope of any one of the industry's professions — mining, mineral processing, geology or economics — and, as a consequence, had not received the attention which it deserved. I could find no authoritative references, only passing mentions in text books and a few papers. It seemed ironic, in an industry devoted to mining ore, that its definition of ore should be so taken for granted. This was over 30 years ago. Since then I have worked on mine planning and economics of cut-off grades for very many mines around the world including most of the large scale operations, both ‘open pit and underground. In the course of this experience, a theo- retical basis for the definition of ore has been established and the theory has been developed to apply to most methods of mining. The concepts are widely accepted and many groups have adopted the techniques as standard practice for the determination of cut-off grade policies. A recent example is Codeleo, the Chilean national copper company. The applications, particularly to the design of big mines, have achieved substantial improvements in the overall economics. Minerals permeate the earth’s crust in varying concentrations around the world. A shovel full of soil from most gardens will probably contain measurable amounts of aluminium, silica, potassium, iron, 2Introduction etc, These may be of interest to the gardener but not normally to the miner for the very obvious reason that the concentrations of the mineral are too low. In fact, concentration is the critical property. The mining industry can be regarded as an industry whose whole concern is with concentration — the progressive concentration of minerals to a form where they become marketable, Typically the process proceeds in stages. The first stage is exploration, which is the search for mineralised regions in the earth’s crust where some degree of natural concentration has already occurred. The second stage is extraction, in which certain parts of a mineralised region are recovered for further treatment. Succeeding stages are treatment stages such as crushing, grinding, flotation, Ieaching, smelting and refining. ‘The subdivision of the whole process into stages, and the locations of the exact boundaries between them, is dependent upon the economics of the technologies involved. For example, finer grinding will usually result in better yields of mineral from the ground material but at the expense of higher power consumption. Similarly, better yields will also usually result from longer residence times in leach circuits but at the expense of higher acid consumption. Thus the stages interact and the optimum combination can only be decided with reference to the whole operation. ‘This book is concerned exclusively with the second stage, extrac- tion. It is often referred to simply as mining, although references to the mining industry are taken to embrace all the stages. The boundary, which is the main focus of attention, is the one which distinguishes the material within a mineralised body that is to be extracted and treated from the remainder. This boundary is com- monly specified by a cut-off grade and the theory which is devel- oped in the following chapters concentrates on such cases. In other cases, where for example blending is necessary to meet some guality requirement, the theory is no longer directly relevant, To avoid ambiguities, the word ore is used solely to describe the material which is extracted for treatment. In other words, by definition, mines extract ore. Hence, establishing an economic basis 3Economic Definition of Ore for determining cut-off grades is, in effect, providing an economic definition of ore. The form of the presentation is first to enunciate the economic principles which are relevant to the analysis and then to trace the consequences of applying these principles in various circumstances. Although this leads to mathematical complexities in some areas, it transpires that clear and consistent thinking is required more often than agility with algebra, During the course of the work a coherent theory of cut-off grades is developed and the derivations of the main formulae arc described in some detail. The intent, however, is to explain the ideas rather than to achieve mathematical rigour. The theory is illustrated with practical case studies which are based upon adaptations of actual applications. Sufficient material is included for the case studies to serve as useful references, At the outset, problems of semantics are encountered. Many important words are commonly used casually in the industry with different meanings in different contexts. This leads to misunder- standings and has certainly inhibited the development of clear concepts, The word “ore” has already been mentioned and is a prime example. Mineralised bodies are called orebodies before they are mined and even before any serious plans to mine them have been mooted. Then reserves of possible, probable, proven, drilled, inferred and developed ore are often quoted when, strictly speaking, they are estimates of tonnages of mineralised material which could be ore under certain circumstances in the future. In this text, care is taken to avoid such inconsistencies of definition. This sometimes Jeads to the unfamiliar usage of certain terms, but the intention is not to be pedantic. Only when a distinction is important are abnormal words and usages adopted. Further problems arise from the finite nature of mineralised bodies. Mining operations based upon them must be of limited duration, a feature which introduces complications into the economic analysis. Unfortunately, it can also introduce an cmotional charge into personal attitudes on the subject. These, aggravated by the semantic difficulties, too often result in unedifying arguments rather than useful discussions. Here, problems of this kind are avoided by 4Introduction the strict adherence to the logical consequences of assumptions and objectives. The aim is the development of a definition of ore that is optimum according to accepted current economic ideas, uncom- promised by other considerations. Optimum economic cut-off grade policies have been calculated for a variety of projects over the past 30 years, Broadly, the conclu sion in the case of mines which are well established is that these optimum policies seidom differ much from current practices. The reason for this is that the mines have usually been designed, and subsequently perhaps modified, to handle the quantities of ore and mineral and the associated grades to which these practices give rise The capacities of the equipment and the installations do not often permit much flexibility and therefore cut-off grades can only be var- ied within narrow limits. In contrast, when expansion schemes are being designed; and even more so when totally new mines are being developed, the theory can indicate cut-off grades quite different from conventional policies with very substantial corresponding improve- ments in the overall returns. Although a surprising number of cut-off grade calculations can be done by hand, they can become very intricate. This observation applies. particularly to the determination of cut-off policies for Jong-range planning. In this book these calculations are performed with a computer program which has been developed for the purpose over the yezrs, It is called OGRE (Optimum Grades for Resource Exploitation). The rights to this program are owned by the Rio Tinto group but other software packages are available. One of the most powerful is OptiCut which is marketed by Whittle Programming in Australia.CHAPTER TWO Economic Principles Inevitably, the question that is asked about any body of mineralisa- tion is— Does it contain‘ore? Or, mote strictly — Does it contain any potential ore? An inconvenient consequence of adopting an economic detinition of ore is that there is no longer any inherent property of the mineralised material which permits an answer to this question in isolation. Although exploration personnel often calculate a ‘dollar value per ton of rock’ in order to assess targets, in fact, mineralsin the ground have no explicit value. Not until they have been extracted, treated and delivered to a customer is any value realised. ‘Therefore, the economics of ore definition cannot be assessed separately from the economics of the total mining process. Indeed, itis the economics of the mining process which determine the economic definition of ore ‘This point is fundamental. Mineralised bodies are often referred to as valuable resources. In a sense they may be but regarding them as such can be misleading. They are certainly not a valuable resource that might be compared with cash in a bank or even a crop on the ground, The only itamediate value they could possess is the price a mining company might bid for the right to mine them. More realistically, a mineralised body should be regarded as a possible opportunity for development, the development being, of course, a mining operation. Any value that might be ascribed to the mineralisa- tion is then realised as an integral part of the proceeds of the ‘operation. It follows from these considerations that, in order to establish an 6Economic Principles economic basis for ore definition, the analysis must first be directed to an operating mine. An understanding of the economic factors which influence ore boundaries must be derived from an under- standing of the economic factors which influence the whole mining process ‘The factors concerned are many and include markets, prices and costs, but they can be integrated using the economic concept of value. A mining operation earns revenue and incurs costs; itis therefore an ‘economic entity and an estimated value can be ascribed to it This value is clearly dependent upon the definition of ore, some ‘vases of definition giving rise to higher values than others. The basis which generates the highest value is optimum and this basis establishes the economic definition of the ore. In other words, material from the mineralised body should be scheduled for mining as ore if, and only if, the decision to treat it adds to the overall economic value of the operation. This is the crucial criterion. Economic value estimates are derived from projected cash flows. Asa result of earning revenues and incurring costs year by year, an operation generates annual net cash flows. These can be amalga- miated into a value, strictly a ‘present’ value, by discounting future flows back at an appropriate cost of capital and totalling them. ‘The theory of present values — ox their inverses, interaal rates of return — and methods for determining the cost of capital are beyond the scope of the present book. The theory has been widely discussed and analysed in many books and papers and it has been accepted in the mining industry, certainly by its financial analysts, It is almost universally used for valuing properties and evaluating new projects. However, its use as a means to determine an optimum operating policy is less common. OF course, this does not make it any the tess valid, but the rity has contributed to scepticism about the present value criterion in this context, particularly when the results differ signifi- cantly from conventional ideas. The differences too, are usually that present value maximisation indicates higher grades and higher rates of mining which seem inconsistent with trusted conservative mining policies.Economic Definition of Ore ‘The expression of reservations about the present value criterion, though, is far from proposing an alternative. In fact, alternatives are rarely formulated unambiguously, but the two most general conten- tions are that: 1) mineralised material should be treated as ore if it will provide a contribution to profit; 2) mining should be conducted in such a way as to maximise the extraction of valuable mineral. ‘The cut-off policies which result from the application of these criteria can be the same, depending upon the definition of the terms employed, but they are discussed separately. The first criterion in some form is popular among technical staff. ‘The question of what constitutes a contribution to profitis the subject ‘of much debate, however. It is often argued that any material for which the value of the recovered mineral will exceed the marginal cost of treating it should be ore. Sometimes a contribution towards overheads is added to the costs and sometimes, beyond this, a minimum profit requirement is also added. The basis of the argument is that if such material is not classified as ore, then an opportunity to earn profit has been wasted. The flaw in the argument is that it totally overlooks capacities. It is equivatent to arguing that a retailer should add to his stock any goods which promise to yield a marginal profit. Retailers do not do this. They are all aware that space is limited and within this limitation they try to stock the more profitable items. Similar considerations apply to a mine. It has a capacity which is limited by some part of the installation — the shaft, the mill, the truck fleet, the rate of development, etc, — and within this limitation it should choose to process the more profitable material. This policy is consistent with the interpretation of the criterion which includes a minimum profit margin, but the supporters of the criterion usually give no basis for determining the margin, other than company policy. The present value criterion, by contrast, gives a precise basis derived as a trade-off between present and future earnings via the present value function. ‘The second criterion that the extraction of valuable mineral should be maximised is frequently proposed by mineral rights owners, local 8Economic Principles governments and conservationists. Of course, it immediately begs a question, what is valuable mineral? An extreme argument is that all the mineral or all the geological reserves (whatever they are) should be extracted in the interests of conserving resources. This is an unrealistic stance which usually stems from a misunderstanding of the way in which minerals are distributed in the ground. A less extreme view is that the mine should be developed in such a way that poorer material is extracted along with richer material in an acceptable blend yielding a satisfactory profit. Of course, every mine blends poorer and richer material of necessity and the point of a cut-off grade is to determine just how poor poorer material can be. The protagonists of the maximum extraction criteria, however, usually imply a degree of subsidy for poor material which would not be economic on its own. What this means is unclear, but the idea of cross-subsidies of ore grades is economically unsound except in special circumstances. A more reasonable view defines valuable material in the same way as in the first criterion. In this case the two give the same result and suffer from the same objection about the effects of capacity. Both of the criteria have another major shortcoming; they do not deal satisfactorily with price variations. Nor do they deal satisfac- torily with variations in other economic parameters, but price is the predominant influence. ‘The point is that, as with all break-even calculations which compare inherent value with cost in some form, higher prices lead to lower cut-off grades. Now, lower cut-off grades yield lower average grades and if the quantity of ore treated remains the same, as it most ‘often does, the output of mineral declines. This is quite the reverse of what should happen in the market. Higher prices imply a deficiency of supply in relation to demand and should prompt an increase in supply, not a decrease. Further, the mine itself is in the position of selling less at the higher price than the lower, a policy which cannot make sense. Present value is the only criterion which does incorporate a means for dealing with varying economic conditions. Parameters defining the conditions are included in the present value estimates and affect the optimum cut-off calculations in a way which avoids nonsensical 9Economie Definition of Ore reactions to price changes. This feature is illustrated in Case Study 2 (p. 112). Many of the criticisms of the present value criterion are actually expressions of special interests. For exampie, it is rarely advan- tageous for the staff of an operating mine to support a policy which shortens its life; they would only be threatening their own liveli- hoods. Similarly, focal governments usually wish to see industrial activities prolonged because this entails continuing employment, continuing taxes and, perhaps, continuing royalties. There is an apparent conflict here between the interests of the contributing parties but, from a strietly economic point of view, it is not an inevitable conflict. If a mine is planned in such a way as to maximise its net present value (excess of present value over capital costs) then, in theory, there is more wealth to share between the participants. Everyone could be better off. Whether, in the event, they are or not depends upon the nature of the agreements between them, but this is a huge subject in its own right To repeat, this book is concerned wholly with the economic definition of ore as that definition which maximises the net present value of a mining operation. There may well be reasons, in special cases, for adopting other bases of definition, necessarily sub- optimum, but such cases are not covered in this book except for ‘occasional incidental references. 10CHAPTER THREE, Finite Resources and Present Values ‘As has been stressed already, every mine is established on a body of mineralisation which is ultimately of limited extent. Some ate very localised and are mined out in a matter of months; others are vast with seemingly endless sources of ore. Nonetheless, they are actually finite and, sooner or later, will be depleted This characteristic makes the analysis of operating strategies for mines very different from the analysis for most other industrial or commercial undertakings. The fundamental concept of optimisation by maximising present values is just as relevant. However, other undertakings are not usually based upon an exhaustable resource and, hence, current operating strategies do not react on the future in the same way. For a mine, higher mining rates will shorten the life and vice versa. The effects of this must somehow be built into the analysis. It is, of course, the present value function itself which provides the means for making cash effects, which occur at different times, commensurate, An essential preliminary to an analysis of cut-off strategy is, therefore, an examination of present value maximisation for an operation based upon a finite resource. This necessarily involves some mathematics but detailed explanations are also given in the following paragraphs. Denote the present value of an operation based upon a finite resource by V. The operation could be a mine but the analysis is gerteral and could apply to any type of finite resource operation like, for example, the liquidation of a stockpile, Vis calculated as the total of the future cash flows discounted back to the present. If these cash flows, year by year, are C,, C2, .. . and iEconomic Definition of Ore the cost of capital is 8 (100 x 6 as a percentage) then V = CU + 6) + Cl + 6)? + The cash flows C,, Co, . . . are dependent on the prices and costs prevailing at the time and therefore the value of V itself is dependent upon the present time, T, which forms the base of the calculation. In other words, the present value of two exactly similar operations calculated at different times will, in general, differ. ie. V = VX) Further, the present value must also depend upon the amount of resource, R, still available. In general, it must decline as the resource is consumed and fall.to zero when the resource is exhausted. ie. V = V(T,R) and V(T,0) = 0 This, however, is not the end of the story. V must depend upon many more variables which describe the way in which the operation is, to be conducted. Rather than writing @ long list, @ convenient mathematical convention is to represent these variables by one symbol. Call it 2. @ defines the operating strategies to be employed in the future and V = V(T.R,Q) In the case of cut-off strategies for 2 mine, @ would consist of the variable cut-off (g, say) which can take on different values, g), 82, gi, - - - forthe remaining years of the mine’s life. Such a sequence of values can be called a policy and therefore gy, 82, 83... . define a cut-off policy, Thus, in this case, if cut-off grades are the only parameters being investigated, Q= 8,8... V = V(T.R,g:.82.-- -) Now, reverting to the general case, of all the sets of operating strategies, 2, which could be adopted, there must be one set, at least, which js optimum in the sense that this set gives rise to the maximum. value for V(T,R,Q). Of, to put it another way, any set of values for 12Finite Resources and Present Values will give rise to some value for V(T,R,2), and if @ is varied over every conceivable set of values. V(T.R,Q) will vary correspondingly, and one, at least, of the values it assumes must be maximum. Call this maximum V*(T,R). A significant observation at this point is that it is no longer a function of ©. VCTR,Q) itself was, but V#(TAR) is not. This is like observing that heights on a map are a function of position but the height of the highest point is not. It has a position, of course, but it is dependent only upon the particular area covered by the map, nothing else. Thus Maxa(V(T,R,Q} = V°(T.R) Before developing the algebra further, consider the function V*(T,R) more closely. It is a function of two variables and therefore forms a surface. For R = 0, when the resource is exhausted, V* = Oso the surface slopes to zero along the R = 0 axis. Also, the function must decrease, in general, as R decreases (only if the optimum strategy involves a cash outflow will it increase) A possible surface is shown in figure 3.1 and an understanding of this figure is important. The surface it represents consistently slopes down towards the lower axis on which the resource available is zero. This axis is itself the zero "present value’ contour. Moving across the figure, the surface has ripples in the time direction. These rip- ples arise from variations in the projections of prices and costs which cause variations in the present value estimates at different times even for the same resource availability. Constant price and cost projections would give a series of horizontal parallel vaiues, Of the two independent variables T and R, the former is not controllable. It is simply the date on which the present value is based, ive, it defines the present to which cash flows are discounted. The latter, R, or more exactly the rate at which R is decremented, is the variable which is directly affected by operations. Ot course, as R is decremented (that is, as resource is consumed) , time necessarily progresses and the changing resource level with time can be graphed, It forms a line or track as is illustrated in figure 3.2. This track defines the rate at which the resource is consumed at every 13Figure 3.1 Maximum Present Value Surface Time Figure 3.2 An Exploitation Track 2020 © “Resource Time 7080 AvailableFinite Resources and Present Values moment throughout the life of the operation and hence it is a complete specification of an exploitation strategy. It can be called an exploitation track. ‘An exploitation track is illustrated in figure 3.2. It is the projection ‘onto the time and resource plane which defines the exploitation strategy. It starts with 1500 units of resource available at the time 2000. About 400 units are consumed in the first 5 years but the rate of exploitation declines to about 100 units in the last 5 years. The resource is exhausted by the year 2035. Every exploitation track fas an associated sequence of present values. Given the projected prices, costs and other economic parameters, the cash flows for each year can be calculated and from them the present values at every point along the track can be compiled. ‘The present values corresponding to the exploitation strategy illus- rated in figure 3.2 are shown as heavy dots. The initial value at the year 2000 is 2500. It declines to 2000 in about year 2007 and then falls steadily to zero by 2035 These present values can be compared with those at the corres- ponding points on the maximum present value surface, whose inter- sections with the two vertical planes are shown on the same figure for reference. Obviously, none can exceed the corresponding maxi- mum and in general they will ali be less. When equality occurs, the remainder of the track from the point of equality onwards defines an ‘optimum strategy track. Moreover, the present values at every subsequent point along such a track must also equal the correspond- jing maxima; if not, an alternative track yielding a higher present value could be constructed in contradiction to the definition of the maximut This principle is fundamental. Aa optimum strategy track must be ‘optimum for every point along its length and all the associated present values must be maxima. In other words, optimum strategy Icacks must be embedded in the maximum present value surface. In general they form a series of parallel tines within the surface, ‘The one corresponding to the same initial position, 1500 units of resource at year 2000, is illustrated in figure 3.3. (Mathematically, IsEconomic Definition of Ore Figure 3.3 An Optimum Exploitation Track on the Maximum Present Value Surface De rttuh teas to 1000 #500 Time ° Resource AvailableFinite Resources and Present Values the application of this principle of optimality is known as dynamic programming). In order to find an optimum operating strategy and the corresponding maximum present value in a particular case, the (olality of the maximum present value surface is irrelevant. Only the surface in the immediate vicinity of the track matters. If some way of deciding the direction of the track at any stage can be found then, given some starting point, the track can be followed and its associ ated present values compiled. This is the purpose of the analysis presented below. Reverting to the algebra and starting with resource R at time T, consider a small decrement of r in R. Let the operating strategy for effecting this decrement be defined by «. Let the time taken be t and let the corresponding cash flow per unit of resource be . t is a function of w and r, and cis a function of w and t ie. t= t(w.r) and c = c(w,t) After taking the decrement, r, the resource remaining will be R-r and the time T+t. This will have a present value V = V(TH.R-1,2') where @ is the operating strategy adopted from this point onwards, The present value at T.R may be denoted by V = V(T.R,ot 2’) because the corresponding operating strategy is w for the decrement r followed by 2’. By definition of present values V(T.Ro+Q’) = re(w,t) + V(THR=1,2'V(1+8)! where 8 is the cost of capital (1008 as a percentage) If is to be an optimum strategy for the decrement r and part of an optimum strategy &, then 2’ must constitute an optimum strategy and V(THt,R=1,) = V*(T+HtR-1) v7Economic Definition of Ore Now, maximising both sides of the equation with respect to @ VECER) = Max (r0(0,0 + V"(T+R-0)/(148))) But assuming r and t are small Va(T+t, R-r) = WA(T,R) + t dV"/dt ~ r dV¥/dr to the first order of approximation and taking the differentials at the point TR. Also Usd) = (48) = 1-8 again to the first order of approximation. Combining these two expressions VeCTHt, R4V(L48) = (1-1) VA(T.R) + t dV A/dt e dvdr neglecting al! the second order terms. Hence, dropping the T,R dependence to simplify the notation V¥=Max,(ro(o,t) + VF - 8t V* +t dVVdT =r dV*/GR} However, V* and therefore dV*/dT and dV#/dR are independent of ©, So also is the arbitrary small increment r. Removing V* and r dV%/dR from the maximisation expression Vie V9 —r dViVdR + Maxe, (re(ont) ~ &V * +t dV/AT) VE cancels so FaVA/dR = Maxy (re(e,t) ~ t[8V* ~ dV#/AT]) or AV*/AR = Maxy (c(t) - [8V * - dV4/T}Ur) [8V* - aV#/dT] is a constant in the maximisation process. Calling it F and putting vr =, the lime taken to process one unit of resource, the equation reduces to AVA/AR = Mary {e609 - Fr) This equation is a quite remarkable result. It implies that dV*/dR must be & maximum at all points along an optimum strategy track. This means that the strategies must be chosen so that every 18Finite Resources and Present Values decrement r in R has the greatest possible effect on V*. This is, intuitively obvious because summing all the changes in V*, corres- ponding to a sequence of decrements in R, all the way to zero will then give the greatest total. Or, in reverse, if an optimum strategy track is followed backwards from R = Oin steps ofr, thea, if each step is taken in the direction of steepest ascent up surface V*, the final point reached must be the highest possible. The remarkable features, however, are on the right-hand side of the equation. Firstly the maximisation is no more difficult than the maximisation of citseif. The only difference introduced by considering the present value of a sequence of cash flows instead of each cash flow individually is the addition of the term Ft, As F is independent of @, this is equivalent to the addition of a time cost term and one term of this nature is almost certainly already in c. Secondly, in order to determine the optimum exploitation strategy atany point, the only information that is required from the remainder of the optimisation process in the futurc is the value of F. A simultaneous maximisation for all the variables in Q is not necessary. ‘This is a vast simplification although, of course, Fis dependent upon the future strategy, s0 some method of determining its value has still to be developed In economic parlance, the term F is an opportunity cost. The operation can be regarded as having a capital value of V. Indeed, this is one interpretation of present value. It is, ina sense, capital which incurs two penalties as a result of being tied up in the operation. One is the interest that could have been earned were it deployed elsewhere; this is 8V, 8 being the corporation's cost of capital. The other is the decline in value as 2 consequence of deteriorating economic conditions; this is —dV*/UT (it can, of course, be 2 bonus father than a penalty if €V*/AT is positive) There are two main alternative methods for determining a value for F in practical applications. The first method is simply to estimate it. This method is satisfactory when the cash flows are not oversensitive to changes in operating strategy and only a current optimum is required. In these circum- 9Economie Definition of Ove stances an approximation for F gives an adequate basis for estab- lishing the optimum because errors in the estimate have little significant effect on the result, Moreover F, however it is determined, is only an economic projection which is subject to 2 measure of uncertainty. This subject is discussed in Chapter 8. ‘The second method is to employ a mathematical iteration tech- nique. A starting vaiue for F is assumed and a complete optimum exploitation strategy calculated, step by step, uatil all the resource is exhausted. The residual present value at this stage should be zero. If itis not, the starting value for F is adjusted and the process repeated. ‘A technique of this kind is well suited to computer applications and has the advantage of deriving a complete exploitation strategy which, in certain cases, can be important, Chapter i0 describes the technique in the context of determining an optimum cut-off grade policy. ‘Summarising, the optimum exploitation strategy for maximising the present value of an operation based upon a finite resource can be determined at any stage by maximising the expression cou 8V*-dV"/dT) = c- Ft...) where cis the cash flow arising from a unit decrement of resource, tis the time taken to process the unit and V* is the maximum present value at that stage. This formula is necessary for the calculation of optimum cut-off grades in conjunction with a suitable economic mod. ‘An important observation to make at this point is that the deriva- tion of formula I does not require any explicit reference to cut-off grides, It js equally relevant to any optimisation where the objective is the maximisation of present value, The opportunity cost measures the trade-off between present and future cash flows and therefore should influence any planning decisions which affect the rate of exploitation and the life of the resource. An example is mill recov ery and throughput of which a brief mention is made in Chapter 27. However, the emphasis in this book is on the cut-off grade decision. 20CHAPTER FOUR Economic Models ‘The purpose of an economic model of an operation is to provide a means for caiculating the effects of changes in certain variables. For present purposes, it is necessary to be able to calculate the effects of changes in cut-off grades, throughputs, prices and costs on the cash flows from a mining operation. Mines differ in many ways and sometimes have unique features so that mine models often have to be specially tailored to fit the circumstances. However, the structure of the mining process is similar in most applications and it is useful to construct a basic model to which variations can be made as necessary. ‘There are three main components to a mining operation related to the throughput with which the component is concerned. The three throughputs are: 1) Mineralised material 2) Ore 3) Mineral ‘The definition of components in this way is fundamental but the interpretation of the definition differs for various types of mine. 1) Mineralised Material ‘This component can be called the mining component. Itis concerned with creating access to the interior of the mineralised body. Costs are incurred per tonne of mineralised material made accessible and capacity is the maximum rate at which the mineralised body can be opened up. In an underground mine this component would normally be called development. It consists of driving headings, raises etc, and forming 21Economic Definition of Ore stopes in preparation for the extraction of ore. In an open pit itis the drilling, blasting and hauling of material since access is gained by total excavation. 2) Ore ‘This can be called the treatment component. It is concerned with the further processing of that part of the mineralised body which is determined to be ore. The costs arc incurred per tonne of ore and the capacity is the maximum throughput of ore that the installation can hhandte. ‘Underground, most of the extraction process, which is normally referred fo as mining, comes into this category because, usually, only the ore is extracted. The remainder of the mineralised body is left in place, Hoisting and.concentrating also belong to this category. In an ‘open pit the definition is clearer. It is the process through which the ore passes after it has been hauied from the pit. ‘The capacity of a mine is normaily quoted as the capacity of the ore treatment component of the operation, For example, a 3,000 tonnes per day mine is understood to be one which can process up to 3,000 tonnes per day of ore. The reason for this convention is that ore handling and concentrating are usually the most capital intensive sections of the mining installations and, more often than any others, they are the final limiting factors on output. 3) Minerat This can be called the marketing component although it might well include smelting and refining as well as selling. The costs are incurred per unit of mineral and the associated capacity is a limit on the output ‘of mineral. Refineries and smelters often impose such a limit. Markets, too, sometimes have an upper bound, ‘This component is essentially the same for all types of mine. Figure 4.1 summarises these definitions graphically. It is important to appreciate that these definitions of the main components in mining do entail significant departures from normal usage in the interpretation of certain words. The reason for thisis the 2Economic Models Figure 4.1 UNDERGROUND SURFACE Developing Drilling and blasting MINING. Raising Loading Cross-cutting Hauling Stoping Tramming Hoisting TREATING Crushing Grinding Separating Smelting MARKETING Refining Selling need to develop a model consistent with both underground and surface applications and to accommodate the three component structure, For example, in an underground operation, stoping and tramming would normally be classified as mining because they are underground activities which are the responsibility of the mining department. Hoisting is more doubtful but it, too, is normally classified as mining In this analysis, however, ali three activities are classified as treatment activities because they are concerned with ore. The point is that material within the mineralised body only incurs stoping, 2BEconomic Definition of Ore tramming and hoisting costs if it is classified as ore; otherwise it is usually left in place and incurs no further cost. Development is normally a clearly recognised activity under- ground. It is the process of creating access to the mineralised body and generally the materia) cannot be adequately sampled to deter- mine its grade and decide whether or not itis to be ore until sufficient access has been gained. Therefore, development costs are incurred per unit of mineralised material opened up. Raising, cross-cutting and often some stope preparation costs aze also essentially, access costs, Here, they are all classified as mining costs for consistency with surface applications. In open pit parlance, stripping is the nearest equivalent to a pure development cost because it is incurred only to gain access to the mineralised body. However, on closet examination it is clear that all mining activities have an access function; even when the material removed is ore, its removal is still essential to reach the material beneath it, Treatment costs are only incurred when material classi- fied as ore starts to attract costs beyond those incurred by waste. Hence, practically all the mining costs are access costs per unit of material within the pit, and this is why the term has been adopted in this text. Mineralised body is another term where the definition requires some care. Itis the total envelope within which miningiis planned. In an open pit context it is everything within the ultimate open pit, Underground it is the whole region within the mining leases to which development is planned to extend. A mineralised body contains mineralised material, by definition, but normally some of it will be of zero grade — for example, overburden and dykes — and only a proportion of the remainder will prove to be economic and hence treated as ore. These definitions have been elaborated at length because they have proved to be a source of some confusion. Most professional miners know what mining costs are by their own definitions and they do not readily adapt to the idea that for cut-off grade determinations such costs should, in certain circumstances, be renamed. Semtantics are really the whole problem. The practicalities are that aEconomic Models the development of a suitable cost model for any mine requires a special study to determine the fixed and variable elements that are relevant for the ranges of variation under study. For cut-off analyses, the variable elements, whatever they are called, must be related to the three throughputs, mineralised material, ore, and mineral. Ingeneral there are substantial fixed elements in what are reported as mining, treatment, or marketing costs and these must be esti- mated, separated and aggregated with the more normally accepted fixed costs such as administration and overheads. Strictly, of course, the word “fixed” in this context is another loose term; such costs vary with time and are more properly called time costs. A convenient notation i Basic Variabie cost Capacity Component Throughput Quantity —hunit threat) (dero'putlyr) Mining Mineralised t m M Material ‘Treating Ore x bh H Marketing Mineral 28. k K where x is the proportion of mineralised material classified as ore and Bis the average grade of the ore as a mineral : ore ratio. Other variables are: & cut-off grade applied to mineralised material to define ore (mineral ote ratio). y yield of mineral in the treatment process (100 y%) F time costs per year (sometimes referred to as fixed costs). P price per unit of mineral, © cash flow per unit of mineralised material. + time taken to work through one unit of mineralised material. Any of these parameters can vary from year to year and this can be represented by quoting the years as suffixes. Other suffixes are also used in the text; their significance is dependent upon the context. Using this notation, the formula for the cash flow arising from one unit of mineralised material is © = (p—K)xyg—xh-m-fr. . . (I) where tis the time taken to progress through the unit. tdepends upon which of the three components of the mining operation is actually 25Economic Definition of Ore limiting throughput. Thus, three cases have now to be examined and this is the subject of the next chapter. Although the majority of the analysis in the remainder of the book is concentrated on the model represented by Expression II, other models can be adopted. The special importance of Expression IL is, that it is the simplest model which reveals the basic revenue and cost structure of a mining operation Complexities often arise in practice from questions related to the allocation of capital costs and the teatment of internal charges and wansfer prices. There are no simple general answers to questions of these kinds, but it is important always to remember the purpose of the exercise. It is, as stated in the opening sentence, to provide a means for calculating the effects of changes in certain variables, and not to satisfy any accounting conventions or managerial objectives. For example, the replacement capital for a fleet of trucks is clearly related to the level of material movements in a pit. It is therefore most realistically incorporated into the mining cost per tonne by calculating an equivalent annual charge and prorating it according to the annual capacities. Another example is an internal smelter charge. If itis the case that an increase in concentrate production would displace concentrate from elsewhere on which the charge could be levied, then the imposition of the charge in a model is realistic. Otherwise it is artificial There ure many variations on these themes, all of which have to be treated on their merits. The objective in every application should always be to develop a model which is faithful to the mine under study. The principles of the subsequent analysis will remain the same. Examples of some different models are given briefly in Chapter 17. 6CHAPTER FIVE Limiting Economic Cut-Off Grades Having derived an expression to be maximised in order to obtain an optimum operating strategy for an undertaking based upon a finite resource (1), and having detived a formula relating cash flows to cut-off grades for a mine (11), it is now possible to combine the two and to quote an expression for determining optimum cut-off grades, Itis: Max,{(p-k)xyg—xh-m-fr—1(8V*-dV*"/dT)} ‘or Max,{(p—k)xyg—xh—m—(F+F)t} For future reference call the expression in brackets v, ise. v = (pok)xyg—xh—-m—(F+F x... (IID) It corresponds to dV*/dR, the rate at which the present value is changing with respect to changes in R, the resource available. In this expression x, the ore to mineralised material ratio, and g, the average grade, are directly dependent upon the cut-off, g. The time +, is also dependent upon g, but indirectly, and three cases must bbe recognised according to the capacity which is limiting output. ‘The three cases give rise to three optimum cut-off grades which are called limiting economic cut-off grades. The data from Case Study 3 (p.114) are used to illustrate the results. The operation concerned is a medium-sized open pit uranium, mine. For ease of reference, the relevant figures are also presented here: Mining variable cost $1-32/tonne of material (m) Treating variable cost $3-41/tonne of ore (h) 2Economic Definition of Ore Mining capacity 12 million tonnestyear (M) Treating capacity 3-9 million tonnes/year (H) Marketing capacity 0-9 thousand tonnes/year (K) Fixed costs S11-9 million/ycar (f) Price net of marketing costs $60 per kilo (p) Recovery 0-87 ) Estimated opportunity cost $15.2 million (F) 1) Mine Limiting With the rate of mining (O/P) or the rate of development (U/G) limiting throughput, the corresponding capacity is M units per year so the time to handle one unit is 1/M. Hence the expression becomes Max, (vm = (p—k)xyg—xh—m—(£+F YM) Only the terms (p~k)xyB—xh vary with gs0 gq, the opt with the mine limiting, is given by Max,(x[(p—k)ya—b]} Here § and x can be represented by integrals of the grade distribution function and the maximum determined by calculus. However, it is possible to see, because of the form of the expression x{(p~k)yg—b}, that g can be lowered for as long as (p—k)yg is greater than h. Hence the breakeven point is given by (P—K)ygm = bh oF Bu = bAp—K)y ‘This is effectively the same breakeven concept that is employed in other determinations of cut-off grades; mineralised material should be classified as ore for as long as its implicit value, (p—k)yg, exceeds. the cost of further processing, h. ‘There are two significant features of the formula for First, it means that the implicit value of mincralised material need only cover the variable cost of treatment (after due allowance for marketing cost, k). Time costs are not relevant and neither is the development nor mining cost. This is so because the formula has been yum cut-off 28Limiting Economic Cut-Off Grades derived on the assumption that the decision to continue operating beyong the current time horizon had already been taken. If it had not, different considerations apply. These are illustrated in Case Study 1.105). Secondly, it does not involve any reference to present values. This means that a mine limited by its mining or development capacity in this way should be operated on a tactical rather than a strategic basis. There is no trade-off of future losses against present gains to modify current policies. For instance, on a tactical basis, if the mineral price changes, the cut-off grade should be changed and in the manner criticised as illogical earlier; namely, as the price rises the grade should be lowered, There is no illogicality in this case because a lower grade increases the proportion of mineralised material classified as ore and hence the total mineral produced; there is no restriction on the treatment capacity to limit the throughput of the extra ore. There- fore, output rises as the price rises. Substituting the data presented above Bm = 341/60 x 0-87 0.07 kiloftonne A cut-off of 0-07 kilo/tonne is actually very low. It arises because if the mine is limiting, the treatment plant nd market are starved of throughput so everything possible is classified as ore, 2) Treatment Limiting, This is the common case with either the ore handling facilities or the concentrating plant restricting throughput. In this case, one unit of mineralised material gives rise to x units of ore and these take x/H units of time to tzeat. Therefore t = x/H and the expression becomes (p—K)xyg—xh—-m—(f+F)x/H) Following the same reasoning as before, this is equivalent to Max,{x[(p—k)y8—h~(E+FY/H]} Max, (vs 29Economic Definition of Ore and g, is given by (p—Wye, = a + (C+ or gy = (h+ (fF FVDM(p— bby ‘This illustrates the point that the present value term F = 8V‘—dV"/dT, appears simply as an additional time cost. ‘This formula is very different from any traditional formula because of the presence of the term F, which can be very significant. Notice that the cut-off grade declines as F declines, which it normally does as the mine ages. This is a particular characteristic of optimum cut-offs determined by the present theory. Again, substituting the data presented above By = GAL + (11-9 + 15-2)3-9) 60 x 0-87 0-20 kilo/tonne ‘This is a much higher cut-off than that obtained when the mine is limiting. Without the opportunity cost, F (=15-2), the formula would give g = B41 + 11-903-9} / 60 x 0-87 0-12 kilo/tonne This figure is probably similar to those derived from traditional cut-off grade analyses which take no account of present value maximisation. ‘The difference illustrates the premium attaching to carlier cash flows when only the treatment facilities are limiting throughput. 3) Market Limiting This can be a genuine market restriction imposed by, say, an exclusive sales contract or it can be the limiting capacity of a refinery or smelter. One unit of mineralised material gives rise to xyg units of mineral which take xyB/K units of time to process or sell. Thus, the optimising expression becomes. (p—K)xy§~xh—m— (+ F)ayB/K) 30 Mare (v4Limiting Economic Cut-Off Grades As before this is equivalent to Max, {x{(p—k— (f+ FY/K)yz—h)) and gy is given by {p-k-(E+F)/K} ye, = b oF & = W{p—k= (I+ FYK}y ‘This and the preceding formulae have in common the property that the fixed costs have been distributed according to the limiting capacity and added to the corresponding variable cost: i.e. for the treatment limiting case, the treatment cost becomes h+ (RM and for the market limiting case, the marketing cost becomes k + (EHEYK ‘The latter formula is novel ia form but it possesses the same typical characteristic that the cut-off grade declines as F declines as the remitting life of the mine is reduced. Applying this formula to the uranium mine data given earlier on page 27 Be = 3-41 / (60-(11-9+15-2)10-9} x 0-87 = 0:13 kilo/tonne ‘This lower figure is a consequence of the fact that a market limitation restricts the extent to which the cash flow pattern can be influenced by the cut-off grade policy. For illustrative purposes, the same value for the opportunity cost hhas been used in all three cases. This is somewhat artificial because the three assumptions about which component is actually limiting the throughput would, if they were applied to the life of the operation, give tise to three different outcomes with correspondingly different opportunity costs, However, the $15:2 million derived in Case Study 3 (p.114) is a typical figure for a mine of this kind and its use gives results which are realistic and comparable. They reveal the profound effects which capacities have on cut-off grade determinations. 31Economic Definition of Ore None of these formulae makes any direct reference to the mineral grades actually present in the mineralised body. Neither do other breakeven formulae. A cut-off grade is calculated with reference solely to costs, prices and capacities quite regardless of the way the grades actually vary within the mineralised body being mined This characteristic, although well accepted within the industry, is nevertheless one which is intuitively unlikely. Surely the cut-off grade ‘must, in general, depend upon the particular grade distribution of the mineralised body under consideration? The answer, in general, is yes and the way in which this actual mineralised grade distribution impinges on the optimum cut-off grade is the subject of the next chapter. However, in the particular situation where only one component of the mining system limits throughput, the factis that the actual grade distribution is not directly relevant. The application of these limiting economic cut-off grade formulae illustrated in the first three Case Studies. ‘The second and third require estimates of present values. This subject is dealt with in Chapter 8 but, for the purpose of the Case Studies, suitable estimates have been obtained in simple but not unrealistic ways. One very important reservation about the application of these formulae and, indeed, about the application of any other breakeven formulae is that they apply to actual grades as they occur in the ground. These are not necessarily identical with the grades as measured for the implementation of a cut-off grade decision. It is a common assumption that they are but itis not one which is generally valid. For example, grade control may be exercised on the basis of one mineral although the value of the ore may be a composite of several minerals. Also there may be significant grade measurement errors. Both of these cause discrepancies between the theoretical boundary between ore and waste given by the breakeven formulae and the actual separation achieved in practice by the imposition of the breakeven grade as an operational cut-off grade. In these circum- stances, the operational cut-off grade is called a parametric cut-off because it is indirectly related to the real grade. Limiting economic cut-off grades still exist but must be calculated with care. This subject, is discussed more fully in Chapter 11. 2