See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/309240217

DETERMINATION OF OPTIMAL MINING CUT-OFF GRADES: MATHEMATICAL

FORMULATION AND SOLUTION ALGORITHM: (CASE STUDY: HINOBA COPPER
MINE IN THE PHILIPPINES)

Thesis · December 2013

DOI: 10.13140/RG.2.2.35292.80007

CITATION READS

1 8,018

1 author:

Daniel Beck
New Mexico Institute of Mining and Technology
2 PUBLICATIONS 4 CITATIONS

SEE PROFILE

All content following this page was uploaded by Daniel Beck on 18 October 2016.

The user has requested enhancement of the downloaded file.

DETERMINATION OF OPTIMAL MINING CUT-OFF GRADES:

MATHEMATICAL FORMULATON AND SOLUTION ALGORITHM

(CASE STUDY: HINOBA COPPER MINE IN THE PHILIPPINES)

Daniel Morgan Boone Beck

Fall 2013

1|Page
 Contents

Chapter: Introduction
1.1 Mining ‘101’
1.2 Definition of Cut-Off Grade
1.3 Calculation of Marginal Cut-Off Grade
1.4 The Importance of Cut-Off Grade

Chapter 2: Literature Review

2.1 Lane’s Seminal Work
2.2 Research Initiatives Post-Lane

Chapter 3: Problem Formulation and Solution

3.1 Lane’s Mathematical Formulation
3.2 Profit Maximization Case – Basic Equations
3.3 Cut-Off Grade for Maximum Profit
3.4 Net Profit Value (NPV) Maximization Case
3.5 An Iterative Solution Algorithm

Chapter 4: Case Study – The Hinoba Copper Project

4.1 Background
4.2 Summary of 2012 Comprehensive Technical Report (CTR)
4.3 The Copper Market
4.4 Model Input Data
4.5 Model Results

Chapter 5: Conclusions and Discussion

2|Page
 Table of Figures

Figure 1: Typical Open Pit Mine Cross Section Showing Economic Ore vs. Waste
Figure 2: Typical Cross Section & Block Model Showing Cu Values Per Block
Figure 3: Example Copper Project: Grade-Tonnage Curve
Figure 4: Example Copper Project: Avg. Grade (Cu%) vs. Cut-Off Grade (Cu%)
Figure 5: Hinoba Project Location
Figure 6: Hinoba Pit Progression
Figure 7: Hinoba Project – Process Flow Sheet
Figure 8: Hinoba Project – Infrastructure and Transportation
Figure 9: Copper Long Term Price Trend
Figure 10: Hinoba Copper Project: Grade-Tonnage Curve
Figure 11: Hinoba Copper Project: Avg. Grade (Cu%) vs. Cut-Off Grade (Cu%)
Figure 12: Hinoba Copper Project: NPV (US$s) vs. Cut-Off Grade (Cu%)

3|Page
 List of Tables

Table 1 – Hinoba Model Input Data

Table 2 – Hinoba NPV vs. Cut-Off Grade (Cu%)

4|Page
 Chapter 1
Introduction
1.1 Mining ‘101’

Mines extract material from natural concentrations of minerals in the Earth’s

crust, called deposits. The distribution of minerals is heterogeneous, implying
that some portions of the deposit will have higher concentrations of minerals
than others. The degree of concentration is referred to as grade. When material
is of sufficiently high grade to be deemed economic, it is referred to as ‘ore’.
Lower grade material is deemed uneconomic and referred to as ‘waste’. The term
cut-off grade is used in the mining industry to distinguish material as either ‘ore’
(i.e. economic) or ‘waste’ (i.e. uneconomic). The challenge is to determine the
cut-off grade that maximizes the value of a deposit and ultimately to mine and
process only ore (i.e. material above the computed cut-off grade) and either leave
in place or discard waste.

Figure 1 – Typical Open Pit Mine Cross Section Showing Economic Ore vs. Waste

Once the cut-off grade is defined (the subject of this paper), the pit design and
mining plan are developed to ensure the optimal delivery of ore to the
concentrator. Mining is typically staged (“cut” or “pushback”) in order to evenly
match the waste movement with the mining and processing of ore. The key input
in determining the mining sequence is a 3D block model of the deposit in which
each block is assigned a grade, tonnage and value. Figure 2 shows a typical cross
section of a block model for an open pit undeveloped copper project. Pit designs
5|Page
and mining sequences are determined by optimization algorithms which consider
the revenue of each block vs. the corresponding cost of its removal. Such
algorithms are beyond the scope of this paper.

Figure 2 – Typical Open Pit Cross Section & Block Model Showing Cu Values Per Block

1.2 Definition of Cut-Off Grade

The most important question in designing and operating a mine is determining

which material is worth mining and processing versus leaving in place or
discarding as waste. The key variable is the concentration of the valuable metal in
the material expressed a percentage of the material containing the metal of
interest. A 1% copper (Cu) grade equates to 10 kilograms of copper (Cu)
contained in every 1,000 kilograms of material. As an example, a calculated cut-
off grade of say 1% Cu in an open pit copper mine means that any material
containing a concentration of more than 1% Cu is economically valuable ore and
worth mining and processing. Any material below the economic cut-off grade is
deemed as waste and the processing of same is considered uneconomic. (A
broader definition of cut-off grade by Taylor, H.K. (1972) is any grade that is used
to separate two courses of action, e.g. to mine or to leave, to mill or to dump,
etc.)

6|Page
1.3 Calculation of Marginal Cut-off Grade

Employing a number of simplifying assumptions (e.g. no capacity constraints, no

fixed costs, no waste, etc.), a marginal cut-off grade can be calculated for any
mineral deposit as follows:

1
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 & 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 ( )
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝐶𝐶𝐶𝐶𝐶𝐶 𝑂𝑂𝑂𝑂𝑂𝑂 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 = 1
𝑡𝑡
𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑋𝑋 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 ( )
𝑡𝑡

The above formula implies that the marginal cut-off grade is the grade at which
the gross value of the ore mined and processed is exactly equal to the cost of
mining and processing. This is equivalent to saying that the cut-off grade is the
grade below which the mineral resource can no longer be processed at a profit.

For the Hinoba Copper Project, which is the subject of a Case Study undertaken in
this paper (Chapter 4), the computed marginal cut-off grade is:

1
$6.25 ( )
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝐶𝐶𝐶𝐶𝐶𝐶 𝑂𝑂𝑂𝑂𝑂𝑂 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 = 𝑡𝑡
1 = 0.11% Cu
83.6% X $6,614 ( )
𝑡𝑡

Hence, any material with a copper (Cu) concentration below the calculated
Marginal Cut-Off grade of 0.11 Cu% is uneconomic and should be left in place or
discarded as waste. Any material with a Cu concentration greater than 0.11 Cu%
is economic ore and worthy of mining and processing. In practice, the optimum
cut-off grade will always be higher than the marginal cut-off grade as the above
calculation understates costs by ignoring fixed G&A costs, stripping costs, capacity
constraints, etc. Nevertheless, it is a useful and simple ‘back-of-the-envelope’
calculation for an initial assessment of the economic potential of a mineral
deposit and is widely used in practice as such.

1.4 The Importance of Cut-Off Grade

7|Page
Mine operators are free to vary cut-off grades, subject to market, technical and
capacity constraints, once operations have commenced. However, doing so has
significant impact on the economics, configuration and life of a mine as it alters
the proportion of the deposit that is deemed economic. Lowering the cut-off
grade increases the proportion of the deposit that is considered economic ore
and therefore targeted for mining and processing. Similarly, increasing the cut-
off grade reduces the proportion of the deposit deemed economic and reduces
the quantity of ore targeted for mining and processing. An illustration of the
trade-off between tonnage (MT) and cut-off grade (Cu%) for the Hinoba Copper
Project is as illustrated in Figure 1 below.
HINOBA GRADE-TONNAGE CURVE
450
400
350
300
Tonnes (Mt)

250
200
150
100
50
0
0.00% 0.10% 0.20% 0.30% 0.40% 0.50% 0.60% 0.70%
Cut-Off Grade (Cu %)

Figure 3 – Example Copper Project: Grade-Tonnage Curve

As the cut-off grade increases, the average grade of the deposit also increases as
illustrated below for the Hinoba Copper Project. As expected, the average grade
of the deposit is always greater than the specified cut-off grade.

8|Page
 Avg. Grade (Cu%) vs. Cut-Off Grade (Cu%)
0.90
0.80
0.70
Avg. Grade (Cu%)

0.60
0.50
0.40
0.30
0.20
0.10
0.00
0.00% 0.10% 0.20% 0.30% 0.40% 0.50% 0.60% 0.70%
Cut-Off Grade (Cu%)

Figure 4 – Example Copper Project: Avg. Grade (Cu%) vs. Cut-Off Grade (Cu%)

A high cut-off grade can be used to boost short term profits, thereby enhancing
near term benefits to stakeholders. However, increasing the cut-off grade is also
likely to decrease the life of the mine. A shorter mine life can reduce time
dependent opportunities, such as those offered by price cycles.

With the possible exception of the underlying commodity price (e.g. copper price
in the case of a copper mine, gold price in the case of a gold mine, etc.), there is
no other single variable that has a greater impact on the economics and
configuration of a mining operation. For the foregoing reason, a detailed analysis
of optimal methods for computation of cut-off grade is a worthwhile endeavor.
Use of a sub-optimal cut-off grade in a mining operation can have severe financial
consequences.

9|Page
 Chapter 2
Literature Review
2.1 Lane’s Seminal Work

Not surprisingly given the importance of cut-off grades to the mining industry,
considerable attention has been devoted to the topic by the mining industry and
independent researchers. The seminal work was undertaken by applied
mathematician Kenneth F. Lane (Choosing the Optimal Cutoff Grade. Colorado
School of Mines Quarterly, Vol. 59, No. 4, 1964, pp. 811-829) and subsequently
expanded in his 1988 book, The Economic Definition of Ore: Cutoff Grades in
Theory and Practice. Lane assumes that a mining operation consists of three
distinct stages (mining, concentrating and refining) and that each stage is subject
to capacity constraints and that the overriding objective is to maximize value as
measured in either Profit Maximization (without regard to the time value of
money) or Net Present Value (NPV). In the case of NPV maximization, he further
expands the concept to that of the opportunity cost of not receiving the future
cash flows earlier due to capacity constraints; i.e. the implied cost of deferring the
processing of higher grade ore as measured by the time value of money (i.e.
discounted cash flows). The opportunity cost can be minimized by operating at
higher cut-off grades and processing higher grade ore sooner. Hence, the cut-off
grade optimization problem is viewed by Lane as a balancing act between
reducing the opportunity costs against a wasting resource, subject to capacity
constraints of each stage of operation. He presents an optimum solution strategy
for the case of Profit Maximization (PM) and an iterative solution algorithm for
the non-linear case of Net Profit Maximization. In practice, the cut-off grade
policy that maximizes NPV is one with declining cut-off grades throughout the life
of the mine. (A more detailed description of these two problem formulations and
associated solution strategies is set out in Chapter 3).

2.2 Research Initiatives Post-Lane

Since the publication of Lane’s original 1964 paper, much work has been devoted
to developing variations of his original optimization algorithm and expanding
10 | P a g e
beyond the classic configuration of a three stage mine producing a single metal
subject to capacity constraints in each stage. Dagdelen (1992, 1993, 1997), and
Whittle et. al. (1995) were amongst the first group of researchers who expanded
on Lane’s original formulation and provided enhanced solution approaches.
Others followed in their footsteps. Ataei et. al. [2003] developed a Golden
Section search method for the determination of an optimum cut-off grade.
Minnitt [2004] presented a streamlined version of Lane’s formulation for a typical
South African Wits style underground gold operation where the operation is
constrained only by mine shaft capacity. Asad [2005, 2007] expanded Lane’s
problem formulation to include multiple metals, price and cost escalation factors.
Bascetin et. al. [2007] introduced a variant of Lane’s solution algorithm
incorporating an optimization factor based on a Reduced Gradient Algorithm
(“GRG”). Rashidinejad et. al. [2008] and Gholamnejad [2008] introduced
environmental cost factors to the basic Lane formulation. Xiao-wei et. al. [2010]
employed Dynamic Programming to optimize cut-off grade in an underground
African copper mine. Abdollahisharif et. al. [2012] presented a further variation
of Lane’s solution strategy to deal with the case of variable capacities. Li et. al.
[2012] proposed a Multistage Stochastic Programming solution approach to the
Lane formulation. Barr [2012] developed a solution strategy (based on finite
element methods and Monte Carlo simulation) that incorporates stochastic metal
prices in lieu of the classic fixed metal price assumption.

All the foregoing research initiatives derive from Lane’s original problem
formulation and solution strategy; i.e. seeking the optimum cut-off grade that
maximizes a mine’s Present Value or Net Present Value subject to capacity
constraints. Given the seminal and enduring nature of Lane’s work, it is the main
focus of this paper and is expanded on in the next Chapter.

11 | P a g e
 Chapter 3
Problem Formulation & Solution

3.1 Lane’s Mathematical Formulation

The determination of economic cut-off grades is a classic Operations Research

optimization problem with a number of variables including profit maximization,
capacity constraints, metal price, operating and fixed costs, etc. The literature
contains numerous publications on determining optimal cut-off grades under
varying assumptions and capacity constraints. In nearly all cases, the principal
objective is to optimize the Net Present Value (NPV) of future cash flows. The
original mathematical formulation of the economic cut-off grade problem, and its
accompanying solution, was set out by Lane (1964) and subsequently expanded in
his 1988 book, The Economic Definition of Ore: Cutoff Grades in Theory and
Practice. Lane’s book sets out a definitive calculus of the NPV criterion using
sequential cash flows from the extraction of mineral reserves from an exhaustible
resource.

Conceptually, Lane assumes that a mining operation is divided into three separate
stages: mining, concentrating and refining. Each stage is assumed to have its
own capacity constraints and costs. Lane then derives formulas for six critical
values for cut-off grade; three limiting cut-off grades and three balanced cut-off
grades. If the mine operates at one of the limiting cut-off grades, it will reach the
capacity of one of the stages. Operating at a balanced cut-off grade ensures that
the capacities of two of the three stages are balanced. Lane demonstrates in his
1964 paper that the optimum cut-off grade is always one of the six derived
solutions.

Lane further explains in his 1988 book that this is really a problem involving
opportunity cost given the capacity constraints of each stage of a mine. To the
extent that lower grade ore is processed today it implies an opportunity cost as
measured by the time value of the deferred profit opportunity associated with
processing a higher grade ore. The opportunity cost can be minimized by
12 | P a g e
processing the highest grade ore soonest subject to the capacity constraints of
the mine. This is equivalent to finding the cut-off grade that maximizes the NPV
of cash flows. Consistent with the foregoing, the cut-off grade policy that
maximizes NPV is one with declining cut-off grades throughout the life of the
mine.

Lane’s formulation of the case of profit (P) maximization, for a single metal,
subject to fixed capacities is as set out below:

MODEL DEFINITIONS

Capacity Constraints
M: Maximum mining capacity in terms of tons per year
C: Maximum concentrator capacity in terms of tons per year
R: Maximum refining/market capacity in terms of tons per year

Costs
m: Mining cost in terms of $ per ton of material mined
c: Concentrating cost in terms of $ per ton of material milled
r: All smelting, refining and marketing costs in $ per ton
f: Fixed costs over the production period (for example, 1 year)

Other
s: Selling price in terms of $ per ton
y: The percentage of the metal recovered in the final product (%)
ĝ: Average grade sent to concentrator
𝑻𝑻: is the production period (for example, 1 year)
𝑄𝑄𝑚𝑚 : Quantity of ore mined (tons)
𝑄𝑄𝑐𝑐 : Quantity of ore sent to concentrator (tons)
𝑄𝑄𝑟𝑟 : Quantity of product produced (tons)

3.2 Profit Maximization Case - Basic Equations

13 | P a g e
Employing the definitions set out above, Lane developed the basic equations as
follows:

Total Costs (𝑇𝑇𝑐𝑐 ):

𝑇𝑇𝑐𝑐 = 𝑚𝑚 ∙ 𝑄𝑄𝑚𝑚 + 𝑐𝑐 ∙ 𝑄𝑄𝑐𝑐 + 𝑟𝑟 ∙ 𝑄𝑄𝑟𝑟 + 𝑓𝑓 ∙ 𝑻𝑻 (1.1)

Since the revenue 𝑹𝑹 is:

𝑹𝑹 = 𝑠𝑠 ∙ 𝑄𝑄𝑟𝑟 (1.2)

The Profit (𝑷𝑷) is given by:

𝑷𝑷 = 𝑹𝑹 − 𝑇𝑇𝑐𝑐 = 𝑠𝑠 ∙ 𝑄𝑄𝑟𝑟 − (𝑚𝑚 ∙ 𝑄𝑄𝑚𝑚 + 𝑐𝑐 ∙ 𝑄𝑄𝑐𝑐 + 𝑟𝑟 ∙ 𝑄𝑄𝑟𝑟 + 𝑓𝑓 ∙ 𝑻𝑻) (1.3)

Combining terms yields the basic profit expression which is the profit from the
next 𝑄𝑄𝑚𝑚 of ore mined:

𝑷𝑷 = (𝑠𝑠 − 𝑟𝑟) ∙ 𝑄𝑄𝑟𝑟 − 𝑚𝑚 ∙ 𝑄𝑄𝑚𝑚 − 𝑐𝑐 ∙ 𝑄𝑄𝑐𝑐 − 𝑓𝑓 ∙ 𝑻𝑻 (1.4)

3.3 Cut-Off Grade for Maximum Profit

As noted by Lane, economic cut-off grade may be limited by mining,

concentrating and/or refining & marketing constraints; consequently six cases
arise depending upon which constraint(s) is (are) limiting factor(s).

Solutions Arising From Constraints of Individual Stages

Determination of the economic cut-off grade using one operation to constrain
total Capacity:

As noted above, the basic profit expression (1.4) is:

P = (s - r) • 𝑄𝑄𝑟𝑟 - m• 𝑄𝑄𝑚𝑚 - c• 𝑄𝑄𝑐𝑐 - f• T

14 | P a g e
Mining Capacity as Limiting Constraint

If mining capacity (𝑀𝑀) is the limiting constraint, then the time needed to mine
material 𝑄𝑄𝑚𝑚 is:

𝑄𝑄𝑚𝑚
𝑇𝑇𝑚𝑚 = (1.5)
𝑀𝑀

Substituting Eq. (1.5) into Eq. (1.4) yields:

𝑓𝑓
𝑷𝑷 = (𝑠𝑠 − 𝑟𝑟) ∙ 𝑄𝑄𝑟𝑟 − 𝑐𝑐 ∙ 𝑄𝑄𝑐𝑐 − (𝑚𝑚 + ) ∙ 𝑄𝑄𝑚𝑚 (1.6)
𝑀𝑀

To determine the grade that maximizes the profit under the mining capacity
constraint, the derivative of Eq. (1.6) must be taken with respect to grade (ĝ):

𝑑𝑑𝑑𝑑 𝑑𝑑𝑄𝑄𝑟𝑟
- c• 𝑑𝑑𝑄𝑄 𝑑𝑑𝑄𝑄𝑚𝑚
𝑓𝑓
= (s - r)• 𝑐𝑐 - (m + )• (1.7)
𝑑𝑑ĝ 𝑑𝑑ĝ 𝑑𝑑ĝ 𝑀𝑀 𝑑𝑑ĝ

Because the amount of material to be mined is independent of the grade, we

have:

𝑑𝑑𝑄𝑄𝑚𝑚
=0 (1.8)
𝑑𝑑ĝ

Therefore, Eq. (1.7) becomes:

𝑑𝑑𝑑𝑑 𝑑𝑑𝑄𝑄𝑟𝑟
= (s - r)•
𝑑𝑑ĝ
- c• 𝑑𝑑𝑄𝑄
𝑑𝑑ĝ
𝑐𝑐 (1.9)
𝑑𝑑ĝ

The amount of product (𝑄𝑄𝑟𝑟 ) is related to the quantity of ore sent to the
concentrator (𝑄𝑄𝑐𝑐 ) by the following relation:
15 | P a g e
𝑄𝑄𝑟𝑟 = ĝ• y•𝑄𝑄𝑐𝑐 (1.10)

Where ĝ is the average grade sent to the concentrator, y is the recovery factor (%)
and 𝑄𝑄𝑐𝑐 is the quantity of ore sent to the concentrator.

𝑑𝑑𝑄𝑄𝑟𝑟 𝑑𝑑𝑄𝑄𝑐𝑐
= = ĝ• y• (1.11)
𝑑𝑑ĝ 𝑑𝑑ĝ

Substituting Eq. (1.11) into Eq. (1.9) yields:

𝑑𝑑𝑑𝑑 𝑑𝑑𝑄𝑄𝑐𝑐
= (s - r)• ĝ • 𝑦𝑦 •
𝑑𝑑ĝ
- c• 𝑑𝑑𝑄𝑄
𝑑𝑑ĝ
𝑐𝑐 (1.12)
𝑑𝑑ĝ

Or

𝑑𝑑𝑑𝑑 𝑑𝑑𝑄𝑄𝑐𝑐
= ⦗(s - r)• ĝ • 𝑦𝑦 − 𝑐𝑐⦘ (1.13)
𝑑𝑑ĝ 𝑑𝑑ĝ

If one sets the derivative of the profit equation with respect to grade equal to
zero, the cut-off grade value that will maximize profit will be the lowest
acceptable value of grade (ĝ) which makes:

𝑑𝑑𝑑𝑑
=0 (1.14)
𝑑𝑑ĝ

Hence, the cut-off grade (ĝ𝑚𝑚 ) considering mining constraints is the value of ĝ
which makes:

(s - r)• ĝ • 𝑦𝑦 − 𝑐𝑐 = 0 (1.15)

16 | P a g e
Thus

𝑐𝑐
ĝ𝑚𝑚 = ĝ = (1.16)
𝑦𝑦•(𝑠𝑠−𝑟𝑟)

Concentrating Capacity as Limiting Constraint

If concentrating capacity (C ) is the limiting constraint, then the time needed to

mine and concentrate material 𝑄𝑄𝑐𝑐 is:

𝑄𝑄𝑐𝑐
𝑇𝑇𝑐𝑐 = (1.17)
𝐶𝐶

Substituting Eq. (1.17) into Eq. (1.4) yields:

𝑄𝑄𝑐𝑐
P = (s - r) • 𝑄𝑄𝑟𝑟 - c• 𝑄𝑄𝑐𝑐 - m• 𝑄𝑄𝑚𝑚 - 𝑓𝑓 • (1.18)
𝐶𝐶

Rearranging terms results in:

𝑓𝑓
P = (s - r) • 𝑄𝑄𝑟𝑟 - m• 𝑄𝑄𝑚𝑚 - (c + )• 𝑄𝑄𝑐𝑐 (1.19)
𝐶𝐶

To determine the grade that maximizes the profit under the concentrating
capacity constraint, the derivative of Eq. (1.19) must be taken with respect to
grade (ĝ):

𝑑𝑑𝑑𝑑 𝑑𝑑𝑄𝑄𝑟𝑟
- m• 𝑑𝑑𝑄𝑄 𝑑𝑑𝑄𝑄𝑐𝑐
𝑓𝑓
= (s - r)• 𝑚𝑚 - (c + )• (1.20)
𝑑𝑑ĝ 𝑑𝑑ĝ 𝑑𝑑ĝ 𝐶𝐶 𝑑𝑑ĝ

As before from Eq. (1.8):

17 | P a g e
𝑑𝑑𝑄𝑄𝑚𝑚
=0
𝑑𝑑ĝ

And from Eq. (1.11)

𝑑𝑑𝑄𝑄𝑟𝑟 𝑑𝑑𝑄𝑄𝑐𝑐
= = ĝ• y•
𝑑𝑑ĝ 𝑑𝑑ĝ

Therefore, Eq. (1.20) becomes:

𝑑𝑑𝑑𝑑 𝑓𝑓 𝑑𝑑𝑄𝑄𝑐𝑐
= ⦗ (s - r)• ĝ • 𝑦𝑦 - (c + ) ⦘• =0 (1.21)
𝑑𝑑ĝ 𝐶𝐶 𝑑𝑑ĝ

Employing the prior logic, the derivative of the profit equation with respect to
grade is set equal to zero, therefore the cut-off grade value that will maximize
profit will be the lowest acceptable value of grade (ĝ) which yields:

𝑓𝑓
(𝑐𝑐 + )
𝐶𝐶
ĝ𝑐𝑐 = ĝ = (1.22)
𝑦𝑦•(𝑠𝑠−𝑟𝑟)

Refining/Marketing Capacity as Limiting Constraint

If refining capacity (R ) is the limiting constraint, then the time needed to mine
and concentrate material 𝑄𝑄𝑟𝑟 is:

𝑄𝑄𝑟𝑟
𝑇𝑇𝑟𝑟 = (1.23)
𝑅𝑅

Substituting Eq. (1.23) into Eq. (1.4) yields:

𝑄𝑄𝑟𝑟
P = (s - r) • 𝑄𝑄𝑟𝑟 - c• 𝑄𝑄𝑐𝑐 - m• 𝑄𝑄𝑚𝑚 - 𝑓𝑓 • (1.24)
𝑅𝑅
18 | P a g e
Rearranging terms results in:

𝑓𝑓
P = (s – r – ) • 𝑄𝑄𝑟𝑟 - c• 𝑄𝑄𝑐𝑐 − m• 𝑄𝑄𝑚𝑚 (1.25)
𝑅𝑅

As before, differentiating with respect to grade (ĝ) and setting the result to zero
yields:

𝑑𝑑𝑑𝑑 𝑑𝑑𝑄𝑄𝑟𝑟 𝑑𝑑𝑄𝑄𝑐𝑐

- m• 𝑑𝑑𝑄𝑄
𝑓𝑓
= (s – r – )• 𝑚𝑚 - c (1.26)
𝑑𝑑ĝ 𝑅𝑅 𝑑𝑑ĝ 𝑑𝑑ĝ 𝑑𝑑ĝ

As before from Eq. (1.8):

𝑑𝑑𝑄𝑄𝑚𝑚
=0
𝑑𝑑ĝ

And from Eq. (1.11)

𝑑𝑑𝑄𝑄𝑟𝑟 𝑑𝑑𝑄𝑄𝑐𝑐
= = ĝ• y•
𝑑𝑑ĝ 𝑑𝑑ĝ

Therefore, Eq. (1.26) becomes:

𝑑𝑑𝑑𝑑 𝑓𝑓 𝑑𝑑𝑄𝑄𝑐𝑐 𝑑𝑑𝑄𝑄𝑐𝑐

= (s – r – )• ĝ • 𝑦𝑦 •
𝑑𝑑ĝ
- c• =0 (1.27)
𝑑𝑑ĝ 𝑅𝑅 𝑑𝑑ĝ

Employing the prior logic, the derivative of the profit equation with respect to
grade is set equal to zero, therefore the cut-off grade value that will maximize
profit will be the lowest acceptable value of grade (ĝ) which yields:

19 | P a g e
 𝑐𝑐
ĝ𝑟𝑟 = ĝ = 𝑓𝑓 (1.28)
(𝑠𝑠 – 𝑟𝑟 – )• 𝑦𝑦
𝑅𝑅

To summarize, for each stage (i.e. Mining-M, Concentrating-C and Refining-R)

subject to its limiting constraint the optimum cut-off grades are defined by
Equations (1.16), (1.22) and (1.28) as follows.

𝑐𝑐
ĝ𝑚𝑚 = ĝ = (1.16)
𝑦𝑦•(𝑠𝑠−𝑟𝑟)

𝑓𝑓
(𝑐𝑐 + )
𝐶𝐶
ĝ𝑐𝑐 = ĝ = (1.22)
𝑦𝑦•(𝑠𝑠−𝑟𝑟)

𝑐𝑐
ĝ𝑟𝑟 = ĝ = 𝑓𝑓 (1.28)
(𝑠𝑠 – 𝑟𝑟 – )• 𝑦𝑦
𝑅𝑅

Observations

The optimum cut-off grade will never be less than ĝ𝑚𝑚 since this is the breakeven
cut-off grade. Similarly, the cut-off grade will never be greater than ĝ𝑐𝑐 since this
would imply consigning economically valuable ore to the waste dump. Hence, the
following relationship holds:

ĝ𝑚𝑚 < ĝ𝑟𝑟 < ĝ𝑐𝑐 (1.29)

Therefore, the optimum cut-off grade (i. e. ĝ𝑜𝑜𝑜𝑜𝑜𝑜 ) that maximizes profit is a value
between ĝ𝑚𝑚 and ĝ𝑐𝑐 , or

ĝ𝑚𝑚 < ĝ𝑜𝑜𝑜𝑜𝑜𝑜 < ĝ𝑐𝑐 (1.30)

20 | P a g e
Solutions Arising from Balancing Stages

In the prior section, we computed optimum cut-off grades for each stage subject
to its capacity constraints. Three optimum cut-off grades for each stage were
derived as set out in Equations (1.16), (1.22) and (1.28). Moreover we have
observed from Eq. (1.30) that the optimum cut-off grade (i. e. ĝ𝑜𝑜𝑜𝑜𝑜𝑜 ) that
maximizes profit is a value between ĝ𝑚𝑚 and ĝ𝑐𝑐 .

As explained by Lane (1988), a second set of cut-off grades need to be calculated

to arrive at a globally optimum solution given the constraints of each stage of the
mining operation. This set of cut-off grade calculations is based on overall
material balance and requires knowledge of the distribution of grades of the
mined material.

For purposes of this calculation, the average grade of the treated material can be
found as a function of the chosen cut-off grade. If two stages are to be balance
(i.e. operate at full capacity), three cases arise.

If both mining and concentrating capacities are the limiting factors, Profit
Equations (1.6) and (1.19) are equal, i.e.

𝑓𝑓 𝑓𝑓
(s - r) • 𝑄𝑄𝑟𝑟 - c• 𝑄𝑄𝑐𝑐 - (m + )• 𝑄𝑄𝑚𝑚 = (s - r) • 𝑄𝑄𝑟𝑟 - m• 𝑄𝑄𝑚𝑚 - (c + )• 𝑄𝑄𝑐𝑐 (1.31)
𝑀𝑀 𝐶𝐶

Equation (1.31) becomes:

𝑄𝑄𝑚𝑚 𝑄𝑄𝑐𝑐
= (1.32)
𝑀𝑀 𝐶𝐶

Thus, the balancing cut-off grade between mine and concentrator (ĝ𝑚𝑚𝑚𝑚 ) is the
cut-off grade that satisfies Eq. (1.32).

21 | P a g e
If both mining and refining capacities are the limiting factors, then Equations (1.6)
and (1.25) are equal resulting in:

𝑄𝑄𝑚𝑚 𝑄𝑄𝑟𝑟
= (1.33)
𝑀𝑀 𝑅𝑅

Thus, the balancing cut-off grade between mine and refinery (ĝ𝑚𝑚𝑚𝑚 ) is the cut-off
grade that satisfies Eq. (1.33).

Finally, in the case where both concentrating and refining capacities are the
limiting factors, then Equations (1.19) and (1.25) are equal resulting in:

𝑄𝑄𝑐𝑐 𝑄𝑄𝑟𝑟
= (1.34)
𝐶𝐶 𝑅𝑅

Thus, the balancing cut-off grade between concentrator and refinery (ĝ𝑐𝑐𝑐𝑐 ) is the
cut-off grade that satisfies Eq. (1.34).

Obtaining a Globally Optimum Solution

At this point, six possible cut-off grades have been calculated. Three
(ĝ𝑚𝑚 , ĝ𝑐𝑐 , & ĝ𝑟𝑟 ) are based capacity limitation of each stage and the other three
(ĝ𝑚𝑚𝑚𝑚 , ĝ𝑚𝑚𝑚𝑚 , & ĝ𝑐𝑐𝑐𝑐 ) are based on the grade distribution of the mined material and
capacities. The objective is to find the cut-off grade which produces the overall
maximum profit considering mining, concentrating and refining constraints. To
find the globally optimum cut-off grade (i. e. ĝ𝑜𝑜𝑜𝑜𝑜𝑜 ) amongst these six possible
solutions, Lane’s method (Lane, 1964) can be applied.

The local optimums for each pair of stages are first considered. The
corresponding optimum grades for each pair (G𝑚𝑚𝑚𝑚 , G𝑚𝑚𝑚𝑚 , & 𝐺𝐺𝑐𝑐𝑐𝑐 ) are selected using
the following rules:

G𝑚𝑚𝑚𝑚 = ĝ𝑚𝑚 , if ĝ𝑚𝑚𝑚𝑚 ≤ ĝ𝑚𝑚 ; or ĝ𝑐𝑐 , if ĝ𝑚𝑚𝑚𝑚 ≥ ĝ𝑐𝑐 ; otherwise ĝ𝑚𝑚𝑚𝑚 (1.35)
22 | P a g e
G𝑚𝑚𝑚𝑚 = ĝ𝑚𝑚 , if ĝ𝑚𝑚𝑚𝑚 ≤ ĝ𝑚𝑚 ; or ĝ𝑟𝑟 , if ĝ𝑚𝑚𝑚𝑚 ≥ ĝ𝑟𝑟 ; otherwise ĝ𝑚𝑚𝑚𝑚 (1.36)

G𝑟𝑟𝑟𝑟 = ĝ𝑟𝑟 , if ĝ𝑟𝑟𝑟𝑟 ≤ ĝ𝑟𝑟 ; or ĝ𝑐𝑐 , if ĝ𝑟𝑟𝑟𝑟 ≥ ĝ𝑐𝑐 ; otherwise ĝ𝑟𝑟𝑟𝑟 (1.37)

The globally optimum cut-off grade (i. e. ĝ𝑜𝑜𝑜𝑜𝑜𝑜 ) is the middle value of
G𝑚𝑚𝑚𝑚 , G𝑚𝑚𝑚𝑚 , & 𝐺𝐺𝑐𝑐𝑐𝑐 , namely:

ĝ𝑜𝑜𝑜𝑜𝑜𝑜 = Middle Value (G𝑚𝑚𝑚𝑚 , G𝑚𝑚𝑚𝑚 ,G𝑐𝑐𝑐𝑐 ) (1.38)

Thus, Equation (1.38) gives the globally optimum calculated cut-off grade that
maximizes profit within the mining, concentrating and refining constraints of the
operation.

3.4 Net Profit Value (NPV) Maximization Case

The previous section demonstrated the selection of optimum cut-off grade,

subject to capacity constraints, to maximize profits. The case of maximization of
Net Present Value (NPV) is an enhanced form of the profit maximization case and
more relevant to real world situations as it explicitly considers the time value of
money.

Profit in year j is defined as 𝑃𝑃𝑗𝑗 .

The total profit (𝑃𝑃𝑇𝑇 ) over the life of mine (n) is, therefore:

𝑃𝑃𝑇𝑇 = ∑𝑛𝑛𝑗𝑗=1 𝑃𝑃𝑗𝑗 (1.39)

The net present value for this uniform series of profits, assuming an annual
interest rate of 𝑖𝑖 is calculated as follows:

(1+𝑖𝑖)𝑛𝑛 −1
𝑁𝑁𝑁𝑁𝑁𝑁 = 𝑃𝑃𝑗𝑗 • ⦗ (1+𝑖𝑖)𝑛𝑛 ] (1.40)
•𝑖𝑖

23 | P a g e
Lane addresses the question of whether NPV can be increased using a cut-off
grade which instead of being fixed varies throughout the mine life. In Lane’s
formulation of the NPV case, he assumes that just prior to mining increment
𝑄𝑄𝑚𝑚 the present value of all remaining profit is V. V is further assumed to be
composed of two parts. The first, 𝑃𝑃𝑃𝑃𝑗𝑗 , is from the profit P realized at time T by
mining the quantity 𝑄𝑄𝑚𝑚 . The second 𝑃𝑃𝑃𝑃𝑤𝑤 is obtained from profits realized by
mining the material remaining after time T. These profits are indicated as 𝑃𝑃𝑗𝑗
occurring at time 𝑇𝑇𝑗𝑗 . The value of all these remaining profits for mining
conducted after t = T as expressed at time T is W. The present values of W and P,
respectively, discounted to time t=o are given by:

𝑊𝑊
𝑃𝑃𝑃𝑃𝑤𝑤 (𝑡𝑡 = 𝑜𝑜) = (1+𝑑𝑑)𝑇𝑇 (1.41)

𝑃𝑃
𝑃𝑃𝑃𝑃𝑝𝑝 (𝑡𝑡 = 𝑜𝑜) = (1+𝑑𝑑)𝑇𝑇 (1.42)

Where d is the discount rate.

The present value at time t = 0 is therefore

𝑊𝑊 𝑃𝑃
V = + (1.43)
(1+𝑑𝑑)𝑇𝑇 (1+𝑑𝑑)𝑇𝑇

Since the present value at time t = T of the remaining reserves is W, the difference
v between the present values of the remaining reserves at times, t=0 and t=T is

v=V–W (1.44)

Equation (1.43) can be rewritten as:

W + P = 𝑉𝑉 • (1 + 𝑑𝑑)𝑇𝑇 (1.45)

24 | P a g e
Applying the binomial expansion to the term (1 + 𝑑𝑑)𝑇𝑇 one finds that

2 23
𝑇𝑇 (𝑇𝑇−1)𝑑𝑑 𝑇𝑇(𝑇𝑇−1)(𝑇𝑇−2)𝑑𝑑
(1 + 𝑑𝑑) = 1 + 𝑇𝑇𝑇𝑇 + 𝑇𝑇 + +⋯ (1.46)
2! 32!

For d small, (1 + 𝑑𝑑)𝑇𝑇 , can be approximated by

(1 + 𝑑𝑑)𝑇𝑇 = 1 + 𝑇𝑇𝑇𝑇 (1.47)

Combining Equations (1.45) and (1.47) results in:

W + P = 𝑉𝑉 • (1 + 𝑑𝑑)𝑇𝑇 = V + VTd

Or

V – W = P – VTd (1.48)

Comparing Equations (1.44) and (1.48) one finds

v = P – VTd (1.49)

The profit P obtained from the mining of 𝑄𝑄𝑚𝑚 in time T is given as before by

𝑃𝑃 = (𝑠𝑠 − 𝑟𝑟) • 𝑄𝑄𝑟𝑟 − 𝑚𝑚 • 𝑄𝑄𝑚𝑚 − 𝑐𝑐 • 𝑄𝑄𝑐𝑐 − 𝑓𝑓 • 𝑇𝑇 (1.50)

Combining Equations (1.49) and (1.50) yields

𝑣𝑣 = (𝑠𝑠 − 𝑟𝑟) • 𝑄𝑄𝑟𝑟 − 𝑚𝑚 • 𝑄𝑄𝑚𝑚 − 𝑐𝑐 • 𝑄𝑄𝑐𝑐 − 𝑇𝑇 • (𝑓𝑓 + 𝑉𝑉𝑉𝑉) (1.51)

The above equation (1.51) must be maximized as one would like to schedule
mining sequences in such a manner that the remaining present values take place
as quickly as feasible. The underlying rationale is that later profits are more

25 | P a g e
heavily discounted than those realized sooner as implied by the time value of
money.

As in the profit maximization case, we first take the derivative of v with respect to
grade and then set the derivative equal to zero. The resulting equations are then
solved for the appropriate cut-off grades subject to mining, concentrating and
refining constraints. This results in the following equations:

𝑐𝑐
ĝ𝑚𝑚 = ĝ = (1.52)
𝑦𝑦•(𝑠𝑠−𝑟𝑟)

𝑓𝑓+𝑑𝑑𝑑𝑑
(𝑐𝑐 + )
𝐶𝐶
ĝ𝑐𝑐 = ĝ = (1.53)
𝑦𝑦•(𝑠𝑠−𝑟𝑟)

𝑐𝑐
ĝ𝑟𝑟 = ĝ = 𝑓𝑓+𝑑𝑑𝑑𝑑 (1.54)
(𝑠𝑠 – 𝑟𝑟 – )• 𝑦𝑦
𝑅𝑅

To summarize, to maximize NPV for each stage (i.e. Mining-M, Concentrating-C

and Refining-R) subject to its limiting constraint, the optimum cut-off grade
calculations are as set out above. These are exactly analogous to the optimum
cut-off grade solutions for the simpler profit maximization case (i.e. Equations
(1.16), (1.22) and (1.28)) except for the added variable V in the calculation of
optimum cut-off grades for ĝ𝑐𝑐 (Eq. 1.53) and ĝ𝑟𝑟 (Eq. 1.54). In both of these
equations, V is of unknown value as it depends on the cut-off grade.

3.5 An Iterative Solution Algorithm

As the cut-off grade depends on the NPV, and the NPV cannot be determined
until the optimum cut-off grades have been found, an iterative solution algorithm
is required. For sake of simplicity, it has been assumed that the concentrator (C)
is the limiting capacity constraint of the mining operation. This is generally the
case in practice for open pit mining operations as the Mining (M) and Refining (R)
26 | P a g e
stages are typically undertaken by third parties and therefore are scalable as
required by Concentrator production levels. This is certainly the case in the
Hinoba Copper Project where the mine configuration is constrained by an
assumed Concentrator capacity of 15 million tonnes per annum (‘tpa’). In the
case of Hinoba, the mining function is being contracted to a third party to
maintain feed to the Concentrator at the rate of 15 million tpa and all
concentrates will be sold to third party refiners and smelters for final processing.

An iterative solution algorithm has been designed, based on Lane’s formulation,

to determine the optimum grades for maximization of NPV. It is summarized as
follows:

Algorithm For Calculation of Cut-Off Grades To Maximize NPV

________________
Step 1
Read Input Files
• Metal price, operating costs, fixed costs
• Recovery rate, discount rate, mine, concentrator and refinery capacities
• Grade & tonnage distribution

Step 2
Set V = 𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖 . The initial 𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖 = 0.

Determine the cut-off grade ĝ𝑚𝑚(𝑖𝑖) by the following equation.

𝑐𝑐+𝑓𝑓+𝐹𝐹𝑖𝑖
ĝ𝑚𝑚(𝑖𝑖) =
𝑦𝑦•(𝑠𝑠−𝑟𝑟)

Where:
𝑐𝑐 is the concentrator cost per ton
𝑓𝑓 is the fixed cost per ton

27 | P a g e
𝐹𝐹𝑖𝑖 is the opportunity cost per ton concentrated in year i
𝑠𝑠 is the selling price per ton
𝑟𝑟 is the refining cost per ton
𝑦𝑦 is recovery of metal expressed as %

Opportunity cost is further defined as:

𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖
𝐹𝐹𝑖𝑖 =
𝐶𝐶

𝑓𝑓𝑎𝑎
𝑓𝑓 =
𝑐𝑐

Where 𝑓𝑓𝑎𝑎 is the annual fixed cost, 𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖 is the NPV of the future cash flows of
the years (i) to the end of mine life (N) and C is the total concentrator capacity in
Year i.

Step 3
Compute the tons of ore 𝑇𝑇𝑂𝑂𝑂𝑂𝑂𝑂 and tons of waste 𝑇𝑇𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 from the grade tonnage
curve.
• The ore tonnage 𝑇𝑇𝑂𝑂𝑂𝑂𝑂𝑂 and the grade ĝ𝑐𝑐 above the cut-off grade ĝ𝑚𝑚(𝑖𝑖)
• The waste tonnage 𝑇𝑇𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 that is below of cut-off grade ĝ𝑚𝑚(𝑖𝑖)
𝑇𝑇𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊
• The stripping ratio (SR) where SR = .
𝑇𝑇𝑂𝑂𝑂𝑂𝑂𝑂

Note: For the Hinoba Case Study, a constant SR of 0.70 is assumed as the SR is
fairly constant throughout the ore body at varying cut-off grade assumptions.

Step 4
Set:
_____
𝑄𝑄𝑐𝑐(𝑖𝑖) = C, if 𝑇𝑇𝑂𝑂𝑂𝑂𝑂𝑂{ ĝ𝑚𝑚(𝑖𝑖)} ≥ C, Otherwise

28 | P a g e
𝑄𝑄𝑐𝑐(𝑖𝑖) = 𝑇𝑇𝑂𝑂𝑂𝑂𝑂𝑂{( ĝ𝑚𝑚(𝑖𝑖)}
_____
𝑄𝑄𝑚𝑚(𝑖𝑖) = 𝑄𝑄𝑐𝑐(𝑖𝑖) •(1 + 𝑆𝑆𝑆𝑆)
_____
𝑄𝑄𝑟𝑟(𝑖𝑖) = 𝑄𝑄𝑐𝑐(𝑖𝑖) • ĝ𝐴𝐴𝐴𝐴𝐴𝐴. •y

Step 5
Compute the annual profit for the life of the mine using a modified version of
Equation 1.50 as follows:

𝑃𝑃𝑖𝑖 = (𝑆𝑆𝑖𝑖 − 𝑟𝑟𝑖𝑖 ) • 𝑄𝑄𝑟𝑟(𝑖𝑖) − 𝑄𝑄𝑐𝑐(𝑖𝑖) • (𝑐𝑐i + 𝑓𝑓) − 𝑚𝑚𝑖𝑖 • 𝑄𝑄𝑚𝑚𝑚𝑚 (1.55)

Step 6
Adjust the grade-tonnage table of the deposit by subtracting ore tons 𝑄𝑄𝑐𝑐(𝑖𝑖) from
the grade distribution intervals above the optimum cut-off grade ĝ𝑚𝑚(𝑖𝑖) and the
waste tons 𝑄𝑄𝑚𝑚(𝑖𝑖) - 𝑄𝑄𝑐𝑐(𝑖𝑖) from the intervals below the optimum cut-off grade ĝ𝑚𝑚(𝑖𝑖)
in proportionate amount such that the distribution is not changed.

Step 7
Check, if 𝑄𝑄𝑐𝑐(𝑖𝑖) is less than the Concentrator capacity, C, then set mine life N=i and
go to Step 8; otherwise set the year indicator to i=i+1 and go to Step 2.

Step 8
Calculate the accumulated future NPVs based on the profits 𝑃𝑃𝑖𝑖 calculated in Step
5 for each year i to N using the following equation.

𝑁𝑁 𝑃𝑃𝑗𝑗
𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖 = � 𝑗𝑗−𝑖𝑖+1 (1.56)
𝑗𝑗=1 (1+𝑑𝑑)

for each year i=1, N where N is the total mine life in years.

29 | P a g e
Note: Given insufficiently detailed block sequence and mine schedule data for
the Hinoba Case Study, it is assumed that equal profits will be achieved from
years 1 to N. The net present value for this uniform series of profits, assuming an
annual interest rate of 𝑖𝑖 is calculated from Equation (1.40), i.e.:

(1+𝑖𝑖)𝑁𝑁 −1
𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖 = 𝑃𝑃𝑗𝑗 • ⦗ (1+𝑖𝑖)𝑁𝑁 ]
•𝑖𝑖

Step 9
Compare the 𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖 with the previous V (Step 2). If the computed 𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖 has not
converged within a preset tolerance of say 5% of V, go to Step 2, otherwise the
program stops. This final calculated cut-off grade is the optimum policy that
maximizes NPV, i.e. ĝ𝑜𝑜𝑜𝑜𝑜𝑜 .
________________________

Comments
Implementation of the algorithm is possible in cases where a continuum of ore-
grade tonnage figures is available along with an ore reserve block model, pit
limits, mine sequence and pushback designs. More specifically, details of each
mining increment, or subdivision of the deposit, are required including the order
in which they are to be mined. Unfortunately, for the Hinoba Case Study only a
single ore grade-tonnage table was available (Table 1) and the underlying block
model, pushbacks, mining sequences, etc. were not available. In addition, in the
case of the Hinoba Case Study, it has been assumed that the concentrator (C) is
the limiting capacity constraint of the mining operation as is typically the situation
in practice. Hinoba’s economic parameters were available including price,
refining cost and metallurgical recoveries as well as operating costs of mining,
concentrating, refining and fixed costs. For the foregoing reasons (i.e. data
limitations) a full implementation of the iterative solution algorithm could not be
executed for the Hinoba Case Study. Instead, a streamlined and simplified model
was developed in Excel to determine and compare NPVs at each of the cut-off
grades provided in the ore grade-tonnage table with the objective of choosing the
optimal cut-off grade that maximizes NPV. Details are discussed in Section 4.4.

30 | P a g e
 Chapter 4
Case Study – The Hinoba Copper Project

This Chapter contains a Case Study for determining the optimum cut-off grade for
the Hinoba Copper Mine in the Philippines using a streamlined and simplified
Excel based solution model.

4.1 Background
Copper Development Corporation (“CDC”, the “Company”) is the holding
company of a group of mineral exploration and development companies. CDC
owns a 92.5 per cent. interest in the Hinoba Copper Project (the “Hinoba
Project”). The Hinoba Project is located on the island of Negros in the Philippines,
approximately 700 km south of Manila. Two known porphyry copper deposits,
the Don Jose (DJ) deposit and the A1 deposit, are contained on the Hinoba
Property.

Figure 5 – Hinoba Project Location

31 | P a g e
Over the years, a significant amount of exploration and metallurgical test work
has been conducted on the Hinoba Project by CDC and its previous owners. To
date, more than US$50 million has been expended on diamond, reverse
circulation and geotechnical drilling, metallurgical test work and engineering
design and cost studies.

4.2 Summary of the 2012 Comprehensive Technical Report (CTR)

CDC commissioned a Comprehensive Technical Report (“CTR”) that was prepared

by independent consultants and includes a resource estimate announced in
October 2011, as well as the following components that are at a pre-feasibility
study (“PFS”) level of accuracy: pit optimization and mining studies (including
costs); metallurgical reports, environmental and social reports; and infrastructure
reports on water, roads, tailings storage facilities and port facilities. The CTR was
delivered in April 2012.

The consultants that prepared these reports are as follows:

32 | P a g e
 Consultant Areas of Responsibility
Aboitz Logistics
Metallurgical test work, Metallurgy, Process and
AMEC Minproc
Study Management
GAIA South Incorporated Socio-Environmental Assessment
Tailings Storage Facility, Infrastructure, Hydrology,
GHD Engineering Water Supply, Communications, Site Buildings and
Camp, Power and Ports
IMC Mining Solutions Pty
Mining Studies and Optimisation
Ltd
Mining Associates Pty
Geology and Resource Estimation
Ltd
Resource Development
Geotechnical, Hydrogeology and Pit Stability
Consultants Ltd.

Resources

The Hinoba Project hosts two known deposits, the Don Jose (DJ) deposit and the
A1 deposit, with an estimated gross JORC-compliant Mineral Resource of 319.3
million tonnes at 0.35% Cu at a 0.2% Cu cut-off grade, which was announced by
the Company in October 2011.

Mining

Pit optimization studies at a US$3.00/lb copper price (i.e. US$6,614/tonne) show

that the DJ deposit contains 138 million tonnes of in-pit mineable ore at a grade
of 0.41% Cu and the A1 deposit contains 88 million tonnes of in-pit mineable ore
at a grade of 0.37% Cu.

33 | P a g e
At the assumed concentrator treatment rate of 15 million tonnes per annum, the
expected mine life of the Hinoba Project is 15.4 years and forecast average annual
production is approximately 46,000 tonnes of copper.

The Hinoba Project will be a conventional open pit operation, with a combination
of bulk and selective mining methods involving a drill, blast, load and haul
operation. Open pit mining will use backhoe excavators and rigid frame dump
trucks to mine and haul the ore.

Figure 6 - Hinoba Pit Progression

Processing

The process flow sheet developed by CDC follows a conventional design with run-
of-mine (ROM) ore delivered to the ROM pad where it will be truck dumped or
34 | P a g e
reclaimed by a front-end loader for feeding into the primary sizer. The primary
crushed product will discharge onto a coarse ore stockpile.
SEMINCO PROJECT SCHEMATIC FLOWSHEET

ROM Ore
Feed
Primary
Crusher
Sizer

Coarse Ore
Stockpile

SAG Mill Ball Mill

Rougher
Flotation

Tailings
Thickeners

Regrind

1st Cleaner
Flotation Tailings
Storage
Facility

2nd Cleaner Cleaner-Scavenger

Flotation Flotation

3rd Cleaner
Flotation

Concentrate
Thickener

Concentrate
Filter

Concentrate Concentrate
Load-out Stockpile

Figure 7 Hinoba Project - Process Flow Sheet

CDC envisages that the crushed ore will be milled in a SAG and ball mill. Following
conventional rougher flotation, the tailings will go out of the plant as the major
part of the reject stream. The rougher concentrate will be reground prior to
cleaning. The cleaner circuit will comprise three sequential stages of closed
circuit cleaners with the tailing from the first cleaner going to a cleaner scavenger.

35 | P a g e
The tailings from the cleaner scavenger will join the rougher tailings and feed to a
thickener before being pumped via pipeline to the tailings storage facility.

The concentrate from the third cleaner stage is the final value product from the
circuit. The concentrate will be thickened and filtered, and then transported by
road to the port site prior to dispatch by sea freight to customers. Dependent
upon head grades, this is forecast to give metallurgical recoveries above 80% and
a concentrate grade assaying approximately 25% copper.

Infrastructure

The Hinoba Project is proposed to be largely self-contained, with mine, mill,

maintenance facilities, administration and fully serviced accommodation camp
located in close proximity to the mine site. The public roads to site will be
upgraded as required to meet demands of extra traffic, particularly concentrate
trucks and freight services. It is proposed that raw water will be extracted from a
fresh water dam close to the site, and a tailings dam will be constructed within 7
km of the mine. The Salvacion Port, which has road access to the mine site, is
recommended as the preferred port for the Project. For power supply to the
Project there are two options. The Company’s preferred source is a geothermal
field on Negros Island that is fully permitted, is currently producing power, and
has the advantages of rapid startup, low generation costs and no greenhouse
gas emissions. The other option is a 100 MW coal-fired power plant on the coast
that several independent power producers have expressed interest in developing
on the basis of an anchor contract for supplying the total 54.8 MW that the
Hinoba Project requires.

36 | P a g e
 Figure 8 Hinoba Project – Infrastructure & Transportation

Financial Analysis

The Company has prepared financial projections based on its view of the most
likely configuration of the Hinoba Project. The Company has estimated initial
capital costs at US$480 million and sustaining capital at US$249 million, i.e. total
capital costs of US$729 million. Average operating costs per pound of payable
copper are forecast at US$1.57 for the life of mine and to average US$1.39 for the
first five years of operation.

37 | P a g e
The Company estimated the Hinoba-an Project, on a gross basis, to have a pre-tax
NPV of US$479 million and a post-tax NPV of US$440 million, assuming a copper
price of US$3.00 per lb (or US$6,614 per tonne) and a debt to equity ratio of
65%:35%. The IRR on post-tax, post-financing cash flow is estimated at 36.3%.
Further details on the basis for the financial projections and the capital and
operating costs are provided in the Company’s presentation on the CTR at the
following link:

http://www.copperdevelopmentcorp.com/investors/presentations.html

The Company has requested that the Author undertake a computation of the
optimum cut-off grade (ĝ𝑜𝑜𝑜𝑜𝑜𝑜 ) to maximize NPV as input to a further set of
economic optimization studies of the Hinoba copper project.

Further information on the Hinoba Project is available from CDC’s website

www.copperdevelopmentcorp.com.

4.3 The Copper Market

Copper has been one of the most important materials in the development of
civilisation and continues to play a vital role in society today. Copper's chemical,
physical and aesthetic properties make it a material of choice in a wide range of
domestic, industrial and high technology applications. Copper is ductile,
corrosion resistant, malleable and an excellent conductor of heat and electricity.
Alloyed with other metals, such as zinc (to form brass), aluminium or tin (to form
bronzes), or nickel, for example, it can acquire new characteristics for use in
highly specialised applications.

Copper is used in building construction, power generation and transmission,

electronic product manufacturing and the production of industrial machinery and
transportation vehicles. Copper wiring and plumbing are integral to the
appliances, heating and cooling systems and telecommunications links used every
day in homes and businesses. Copper is an essential component in the motors,
38 | P a g e
wiring, radiators, connectors, brakes and bearings used in cars and trucks. The
average car contains 1.5 km (0.9 mile) of copper wire, and the total amount of
copper ranges from 20 kilograms (44 pounds) in small cars to 45 kilograms (99
pounds) in luxury and hybrid vehicles.

One of copper’s more recent applications includes its use in frequently touched
surfaces (such as brass doorknobs), where copper’s antimicrobial properties
reduce the transfer of germs and disease. Semiconductor manufacturers have
also begun using copper for circuitry in silicon chips, which enables
microprocessors to operate faster and use less energy. Copper rotors have also
recently been found to increase the efficiency of electric motors, which are a
major consumer of electric power.

Real Copper Prices, 1900 to 2010

(2009 US$/lb)

$6.00
$5.00
$4.00
$3.00
$2.00
$1.00
$0.00
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010

Figure 9 – Copper Long Term Price Trend

The world’s production and consumption of copper have increased dramatically in

the past 25 years. As large developing countries have entered the global market,
demand for mineral commodities, including copper, has increased. In the past 20
years, the Andean region of South America has emerged as the world’s most

39 | P a g e
productive copper region. In 2011, approximately 42 per cent. of the world’s
copper was produced from the Andes Mountains region.

The United States was the leading copper consumer until 2002 when it was
overtaken by China as the world’s leading user of refined copper. The booming
economy in China contributed to a tripling of its annual refined copper
consumption during the eight years from 1999 to 2007.

4.4 Model Input Data

The following model input data has been derived from consultations with CDC
and review of the historic feasibility and engineering work, especially the Hinoba
Comprehensive Technical Report (CTR):

Table 1 – Hinoba Model Input Parameters

HINOBA COPPER PROJECT
CUTOFF GRADE CALCULATION
INPUT DATA

Copper Price (US$/tonne) s 6,614

Concentrator Capacity (Mtpa) C 15

Mining Capacity (tpa)
Refining Capacity (tpa)
Fixed Costs (US$/tonne) - Ore
Concentrator Costs (US$/tonne)
Mining Costs (US$/tonne)
M Unlimited
R Unlimited
f 0.51
c 5.49
m 2.00

Refining Costs (US$/tonne) r 1,080

Average Grade Deposit (% Cu) ĝ-Avg 0.36

Recovery (%) y 0.836
Discount/Interest Rate (%) d or i 0.10

40 | P a g e
Ore-Tonnage Cut-Off Table
Tonnes
COG (Mt) Cu %

0% 411 0.30

0.10% 403 0.31

0.20% 292 0.36

0.30% 138 0.50

0.40% 123 0.52

0.50% 63 0.59

0.60% 23 0.68

0.70% 7 0.78

The trade-off between ore tonnage and cut-off grade (Cu%) for the Hinoba
Copper Project is as shown below.

HINOBA GRADE-TONNAGE CURVE

450
400
350
300
Tonnes (Mt)

250
200
150
100
50
0
0.00% 0.10% 0.20% 0.30% 0.40% 0.50% 0.60% 0.70%
Cut-Off Grade (Cu %)

Figure 10 – Hinoba Copper Project: Grade-Tonnage Curve

Similarly, the relationship between the average grade (Cu%) and cut-off grade
(Cu%) is as shown below.

41 | P a g e
 Avg. Grade (Cu%) vs. Cut-Off Grade (Cu%)
0.90
0.80
0.70
Avg. Grade (Cu%)

0.60
0.50
0.40
0.30
0.20
0.10
0.00
0.00% 0.10% 0.20% 0.30% 0.40% 0.50% 0.60% 0.70%
Cut-Off Grade (Cu%)

Figure 11 – Hinoba Copper Project: Avg. Grade (Cu%) vs. Cut-Off Grade (Cu%)

The majority of the input data was taken directly from the Company’s
Comprehensive Technical Report (CTR) published in April 2012. Some parameters
had to be estimated and simplified as the underlying Block model, pushback
sequence and mine schedule were not available to the author. In addition, a
detailed ore-grade tonnage table for each mining increment was not available.
The only ore-grade tonnage data available was that set out in Table 1 which
provided tonnage estimates for eight cut-off grades only (i.e,. 0.0 Cu%, 0.1 Cu%,
0.2 Cu% … 0.7 Cu%).

As a result of the foregoing, the parameters estimated and/or simplified include:

• Use of a constant Strip Ratio of (0.70). We believe this to be a reasonable

approximation as the Strip Ratio is fairly constant throughout the ore
deposit.

• The mine was modeled as a single pushback with one ore grade-tonnage
curve throughout the mine life (see Grade-Tonnage figures in Table 1) as
this was the only data available.

42 | P a g e
 • Per the Company’s design specifications, the Concentrator capacity (C) was
fixed at 15 million tonnes per annum and the capacity of the mining and
refining stages were assumed to be unlimited as both will be contracted
out to third parties. This is fairly standard practice in the mining industry.
For purposes of the model, mining was maintained at a rate to keep the
Concentrator operating at its full design capacity of 15 million tonnes per
annum. Other capacity constraints were ignored.

• Given insufficiently detailed block sequence and mine schedule data for
the Hinoba Case Study, it was assumed that constant grade and equal
profits will be achieved from years 1 to N.

Due to the above noted input data limitations, and associated simplifications, the
Iterative Solution Algorithm set out in Section 3.5 could not be implemented.
Instead, a streamlined and simplified Excel spreadsheet model was prepared in
which NPVs were calculated for each of the eight cut-off grades set out in Table 1
with the objective of choosing the optimum cut-off grade that maximizes the
project’s NPV. The result should be approximately the same as that which would
be derived by the Iterative Solution Algorithm if all the requisite input data were
available.

4.5 Model Results

The calculated NPVs for each of the cut-off grades are as shown in Table 2 below.
A graph of results is shown in Figure 11 below.

Table 2 – Hinoba: Cut-Off Grade (CU%) Vs. NPV

NPV Mine Life

Cut-off Grade (US$ Million) (Years)
0.00 Cu% 622 27
0.10 Cu% 683 26
0.20 Cu% 918 19

43 | P a g e
0.30 Cu% 1,204 9
0.40 Cu% 1,193 8
0.50 Cu% 881 4
0.60 Cu% 441 2
0.70 Cu% 169 0.5

HINOBA OPTIMIZATION - NPV ($) VS. CUT-OFF GRADE

$1,400,000,000
$1,200,000,000
NET PRESENT VALUE (US$)

$1,000,000,000
$800,000,000
$600,000,000
$400,000,000
$200,000,000
$-
0.00% 0.10% 0.20% 0.30% 0.40% 0.050% 0.60% 0.70%
CUT-OFF GRADE (CU%)

Figure 12 – Hinoba Copper Project: NPV (US$s) vs. Cut-Off Grade (Cu%)

As noted Figure 12 above, NPV increases with cut-off grade with a maximum NPV
of US$1.2 billion realized at a cut-off grade of 0.30 Cu%. At higher cut-off grades,
the NPV deteriorates reflecting lost profits from the increasingly shorter mine life.
As explained in Chapter 1, increasing cut-off grade improves near term
profitability but at the cost of mine life. In the case of Hinoba, the optimum
tipping point between improved profitability and reduced mine life is at a cut-off
grade of 0.30 Cu%.

Deducting Hinoba’s development and sustaining capital costs of US$729 million

from the above calculated NPV of future profit streams (i.e. US$1.2 billion @ 0.30
Cu%), yields a pre-tax NPV estimate of US$475 million. This figure is almost
identical to the US$479 million pre-tax NPV estimate contained in the Hinoba

44 | P a g e
Comprehensive Technical Report and serves to validate the model. (See Section
4.2 – Financial Analysis for a summary of the Hinoba CTR findings.)

The Hinoba CTR assumes a cut-off grade of 0.20 Cu% while our analysis suggests
the optimum cut-off grade should be set at a higher level of 0.30 Cu%. This
change would decrease the mine life from 15 to 9 years and improve the project’s
NPV. While we were forced to make a number of simplifying assumptions due to
data limitations, our model suggests it worthwhile for CDC to undertake further
optimization studies using a cut-off grade of 0.30 Cu%. The NPV improvements
could be in the tens of millions of dollars, most definitely worthy of additional
analysis. We also recommend that any further analysis consider the computation
of a second marginal cut-off grade for purposes of stockpiling lower grade ore for
potential future processing in the event of higher copper prices.

45 | P a g e
 Chapter 5
Conclusions & Discussion

In the mining industry, cut-off grade is arguably the single most important policy
and operating parameter. Its value directly impacts the economics, configuration
and life of a mine as it alters the proportion of the deposit that is deemed
economic. As such, a detailed analysis of optimal methods for the computation of
cut-off grade is a worthwhile endeavor. Use of a sub-optimal cut-off grade can
have severe financial consequences.

In this paper we present Lane’s (1964,1988) formulation of the determination of

economic cut-off grades as a classic Operations Research optimization problem
with a number of variables including profit maximization, capacity constraints,
metal price, operating and fixed costs, etc. We enumerate the underlying
mathematics of Lane’s formulation and show that in the case of maximization of
Net Present Value (NPV) it is a non-linear problem as the determination of NPV
depends on cut-off grade and cut-off grade in turn is a function of NPV. We
present an iterative solution algorithm based on Lane’s problem formulation and
solution strategy.

Finally, we undertake a determination of the optimum cut-off grade for the

Hinoba Copper Project, a planned US$480 million open-pit copper development
project in the Philippines. Based on data provided by the project’s owner, we
implement an Excel based solution and determine the optimum cut-off grade to
be 0.30 Cu%. We compare our results with those contained in the Company’s
Comprehensive Technical Report and note that a higher NPV can be achieved
using our calculated 0.30 Cu% cut-off grade in lieu of the report’s assumed cut-off
grade of 0.20 Cu%. We recommend that further optimization studies be
undertaken by the Company using our recommended cut-off grade of 0.30 Cu% as
the financial benefit could be in the tens of millions of dollars. We also
recommend that such further studies consider the economics of stockpiling lower
grade ore for future processing in the event of higher copper prices.

46 | P a g e
In terms of future research there is much to be done. More efficient solution
strategies are required to cope with real world mining projects involving variables
and constraints beyond the classic Lane formulation, e.g. stockpiling options,
multiple metals, environmental costs, stochastic inputs, integration with mine
schedules, pit optimization, equipment utilization, etc. Progress has been made
but more work is required and justified given the sizable potential financial
rewards for the mining industry. Newman et. al. [2010] suggest a number of
areas worthy of future research.

47 | P a g e
 References

Abdollahisharif, J, Bakhtavar, E. and Anemangely, M., Optimal Cut-Off Grade

Determination Based on Variable Capacities in Open-Pit Mining, The Journal of the
South African Institute of Mining and Metallurgy, December 2012, Vol. 112, 1065-
1069

Asad, M.W. A., Cutoff Grade Optimization Algorithm with Stockpiling Option for
Open Pit Mining Operations of Two Economic Metals, International Journal of
Surface Mining, Reclamation and Environment, Vo. 19, No. 3, September 2005,
pp. 176-187

Asad, M.W.A., Optimum Cut-Off Grade Policy for Open Pit Mining Operations
Through Newt Present Value Algorithm Considering Metal Price and Cost
Escalation, Engineering Computations: International Journal for Computer-Aided
Engineering and Software, Vol. 24, No. 7, 2007, pp. 723-736

Ataei, M. and Osanloo, M, Determination of Optimum Cutoff Grades of Multiple

Metal Deposits By Using the Golden Section Search Method, The Journal of the
South African Institute of Mining and Metallurgy, October 2003, pp. 493-500

Barr, D., Stochastic Dynamic Optimization of Cut-Off Grade in Open Pit Mines,
Queen’s University, April 2012, Master’s Thesis

Bascetin, A. and Nieto, A., Determination of Optimal Cut-Off Grade Policy to

Optimize NPV Using a New Approach with Optimization Factor, The Journal of the
South African Institute of Mining and Metallurgy, February 2007, Vol. 107, pp. 87-
94

Copper Development Corporation, Hinoba Copper Project, Comprehensive

Technical Report (CTR), April, 2012

48 | P a g e
Dagdelen, K., Cut-off Grade Optimization, 23rd Application of Computers and
Operations Research in the Mineral Industry, SME, Littleton, Colorado, USA, 1992,
pp. 157-165

Dagdelen, K., An NPV Optimization Algorithm for Open-Pit Mine Design, 24th
Application of Computers and Operations Research in the Mineral Industry, SME,
Littleton, Montreal, Quebec, 1993, pp. 257-263

Dagdelen, K. and Mohammad, W.A., Multi-Mineral Cut-Off Grade Optimization

with Option to Stockpile, 1997, SME Annual Meeting, Denver, Colorado

Eksioglu, S.D., Seref, M.M.H., Ahuja, R. K., Winston, W. L., Developing

Spreadsheed-Based Decision Support Systems, 2nd Edition, 2011

Hustrulid, W., Kuchta, M. and Martin, R., Open Pit Mine Planning & Design, CRC
Press, 3rd Edition, 2013

Gholamnejad, J., Determination of the Optimum Cutoff Grade Considering

Environmental Cost, Journal of International Environmental Application & Science,
2008, Vol. 3(3), pp. 186-194

Lane, K.F., Choosing the Optimum Cutoff Grade, Colorado School of Mines
Quarterly, 1964, Vol. 59, pp. 811-829

Lane, K.F., The Economic Definition of Ore, Cutoff Grade in Theory and Practice,
Mining Journal Books, London, 2008

Li, S. and Yang, C., An Optimum Algorithm for Cut-Off Calculation Using Multistage
Stochastic Programming, Journal of Theoretical and Applied Information
Technology, November 2012, Vol. 45. No. 1, pp. 117-122

49 | P a g e
 Minnitt, R.C.A., Cut-off Grade Determination for the Maximum Value of a Small
Wits-type Gold Mining Operation, The Journal of the South African Institute of
Mining and Metallurgy, June 2004, pp. 277-284

Newman, A., Rubio, E., Caro, R., Weintraub, A. Eurek, K., A Review of Operations
Research in Mine Planning, Interfaces, Vol. 40, No. 3, May-June 2010, pp. 222-245

Rashidinejad, F, Osanloo, M and Rezai, B., An Environmental Oriented Model for

Optimal Cut-off Grades in Open Pit Mining Projects to Minimize Acid Mine
Drainage, International Journal of Environmental Science and Technology, Spring
2008, Vol. 5, pp. 183-194

Rendu, J.M., An Introduction to Cut-Off Grade Estimation, 2008, Society for

Mining, Metallurgy and Exploration Inc.

Taylor, H.K., General Background Theory of Cut-Off Grades, Institution of Mining

and Metallurgy Transactions, 1972, A160-179

Whittle, J. and Wharton, C.L., Optimizing Cut-Offs Over Time, Proceedings of 25th
International Symposium on Application of Computers and Mathematics in the
Mineral Industries (APCOM), Brisbane, Australia, 1995, pp. 261-265

Xiao-wei, G., Wang, Q., Chu, D. & Zhang, B., Dynamic Optimization of Cutoff Grade
in Underground Metal Mining, 2010, J. Cent. South Univ. Technology, pp. 492-497

Yi., R. and Sturgul, J., Analysis of Cutoff Grades Using Optimum Control,
Proceedings of the Twentieth International Symposium on the Application of
Computers and Mathematics in the Minerals Industries, 1988, pp. 263-269, The
Journal of the South African Institute of Mining and Metallurgy.

50 | P a g e

View publication stats