HOME

A Cut-Off of Liberated and Selected Ore

Minerals Optimisation Based on the
Geometallurgy Concept
G Turner-Saad1

ABSTRACT
An improvement to cut-off grade optimisation theory based on geometallurgy has been completed.
The improvement fundamentally consisted of taking into account mineralogical and textural
characteristics instead of grades. These characteristics are related to the spatial variability of
mineral abundance, association, particle size and liberation properties. Additionally, the improve-
ment considered the spatial variability of mineral processing liberation and selectivity properties
associated with mineralogy and texture. The enhanced optimisation enables estimation of the
economic value of mining operations based on an optimum cut-off policy attributed to liberated
and selected ore minerals.
The enhanced optimisation was developed as an essential component of a joint cut-and-fill
mining and mineral processing methodology based on mixed integer mathematical programming.
The formulation considers static and dynamic modifying factors that vary over the projected life of
mining. The objective function of the mathematical formulation consists of maximising the realistic
expected economic value of concentrates or products of liberated and selected ore minerals,
whilst minimises liberated and selected gangue minerals. The optimal solution is obtained when
the objective function is subject to geological, mining, processing, marketing, smelting, refining,
environmental and financial constraints. The methodology accesses geometallurgical multivariate
resource models, which integrate the spatial variability of mineralogical and textural characteristics
and mineral processing liberation and selectivity properties.
The geometallurgical process concurrently optimises stope geometries, ore reserves, mining
sequences, and mining and mineral processing production based on the enhanced cut-off of
liberated and selected ore minerals. In addition, the process considers additive and non-additive
transfer functions associated to mutually exclusive geometallurgical spatial domains. The transfer
functions also take into consideration the blending of mineralogical and textural characteristics
with mineral processing liberation and selectivity properties.
This contribution presents an example of cut-off grade optimisation via mineralogical and
textural characteristic, and liberation and selectivity processing parameter optimisation for a
spatially variable orebody.

INTRODUCTION
The most important objective of the strategic mine planning However, a realistic expected economic value can be
process consists in maximising the expected economic determined by considering the spatial variability of mineral-
or net present value (Davis and Newman, 2008). This is ogical and textural characteristics and associated liberation
achieved by mining and processing the ore reserves, and and selectivity properties (Turner-Saad, 2010). In addition,
the marketing, smelting and refining of the concentrates or reliable static and dynamic assumptions can be made of the
products over the projected mine life. The maximisation of the modifying factors over the projected mine life.
expected economic or net present value can be determined by The straightforward principle of determining the expected
applying current state-of-the-art cut-off grade optimisation economic value of mining, processing and marketing the
methodologies (Lane, 1964, 1979, 1988; Rudenno, 1979; concentrates or products of two independent discrete ore
Lane et al, 1984; Dagdelen,1993). These methodologies are reserve blocks is described. The most important assumption
fundamentally based on the spatial distribution of orebody is that the two ore reserve blocks with the same volume,
grades. In addition, account can be made of reasonable static bulk density, grade, dilution and static mineral processing
and dynamic assumptions on some of the modifying factors in recovery will produce the same recovered metal content.
converting mineral resources to ore reserves (eg JORC, 2004; Nevertheless it is likely that the real recovered metal content
Napier, 1983; Baird and Satchwell, 2001; Asad, 2005). of these two discrete blocks will be different. This is due to the

1. Executive Geometallurgical Consultant, CAE Mining, Level 23, 333 Ann Street, Brisbane Qld 4000. Email: gts@datamine.co.uk

THE FIRST AUSIMM INTERNATIONAL GEOMETALLURGY CONFERENCE / BRISBANE, QLD, 5 - 7 SEPTEMBER 2011 263
G TURNER-SAAD

difference in the mineralogical and textural characteristics, variability of mineralogical and textural characteristics, and
and mineral processing performance of the two blocks. In liberation and selectivity mineral processing properties of an
general terms, the mineral processing recovery is a function orebody.
of the abundance, association, grain size and liberation The purpose of the proposed methodology consists of assess-
characteristics and product size properties of ore and gangue ing the realistic expected economic value of mining operations
minerals. It is convenient to emphasise that concentrates or by defining ‘what if’ strategic, tactical and operational
products are constituted by liberated and selected ore and scenarios. The assessment can be performed by defining
gangue minerals. and varying geological, mining, processing, marketing,
On other hand, underground mine optimisation method- smelting, refining, environmental and financial scenarios and
ologies relate to stope geometries, mining sequences and assumptions over the projected mine life.
mining production have been developed as isolated stages
by a number of researchers as is shown in Figure 1 (Murray METHODOLOGY
and Magri, 1978; Trout and Grice, 1993; Alford, 1995; Ovanic
In this study, the objective of the geometallurgical optimisation
and Young, 1995; Ataee-pour and Baafi, 1998; Van Leuven,
process is to maximise the expected economic value of under-
1998; Thomas and Earl, 1999; Rahal et al, 2003; Smith,
ground cut-and-fill mining and mineral processing operations.
Sheppard and Karunatillake, 2003; Topal, Kuchta and
The estimation of expected economic value throughout
Newman, 2003; Grieco, 2004; Smith and O’Rourke, 2005).
These mathematical formulations did not take into account this methodology is based a function of static and dynamic
the combined relationship of the underground mining technical and financial factors over the projected mine life.
stages either for single or multiple commodity deposits. The essence of the methodology resides in defining optimum:
Additionally, there are no publications on the integration of  mining, blending, stockpiling and processing of reserves
the geometallurgical concept, and specifically for underground with different mineralogical and textural characteristics,
cut-and-fill mining. and mineral processing liberation and selectivity proper-
ties; and
 marketing, smelting and refining concentrates or products
Mineral Geometallurgical with different liberated and selected ore and gangue
Resources Resources
minerals,
where optimisation is in agreement with geological, mining,
Mining Mining processing, marketing, smelting, refining, environmental and
Design Design financial constraints.
The development of the geometallurgical optimisation
process is based on five activities, inclusive of:
Ore Ore
Reserves Reserves 1. accessing geometallurgical multivariate resource models
of a deposit;
2. defining the strategic, tactical and operational scenarios;
Mining Mining 3. defining the static and dynamic modifying factors over the
Sequence Sequence
projected mine;
4. optimising simultaneously stope geometries, ore reserves,
Mining Mining mining sequences, mining and mineral processing
Production Production productions; and
5. assessing the realistic expected economic value.
Optimisation is an iterative process due to the unlimited
Processing Processing
number of ‘what if’ strategic, tactical and operational
Production Production
scenarios that can be considered and because of the unlimited
number of assumptions of the modifying factors.
Smelting Smelting The iterative optimisation process is based on concurrent
Production Production integration of stope geometries, ore reserves, mining
sequences, mining and mineral processing productions over
the projected mine life as is shown in process diagram of
Refining Refining Figure 1.
Production Production
The methodology is based on a mixed integer mathematical
programming formulation developed in AMPL A Mathematical
FIG 1 - Process diagrams of the isolated (left) and concurrent Programming Language (Fourer, Gay and Kernigham, 1993)
(right) mining and mineral processing optimisation. and using the IBM ILOG CPLEX Optimizer solver (IBM, 2011).
A summary of the objective function and geological, mining,
A geometallurgically integrated and iterative underground processing, marketing, smelting, refining, environmental and
cut-and-fill mining and mineral processing optimisation financial constraints of the mathematical formulation are
methodology based on a mixed integer mathematical program- described in the next sections.
ming has been developed. The methodology described is
an extension and enhancement of the research previously Objective function
published by Turner-Saad and Smith (2006) and Turner- The objective function consists in maximising the realistic
Saad (2011), which integrates geometallurgical multivariate expected economic or net present value of concentrates
resources models of mineral deposits. This extended and or products of liberated and selected ore minerals whilst
enhanced methodology takes into consideration the spatial minimising liberated and selected gangue minerals.

264 THE FIRST AUSIMM INTERNATIONAL GEOMETALLURGY CONFERENCE / BRISBANE, QLD, 5 - 7 SEPTEMBER 2011
 A CUT-OFF OF LIBERATED AND SELECTED ORE MINERALS OPTIMISATION BASED ON THE GEOMETALLURGY CONCEPT

The maximisation is controlled by a cut-off of liberated and on whether the mineral processing recovery is treated as a
selected ore minerals optimisation methodology, which it is static, dynamic additive or non-additive transfer function.
an adaptation and enhancement of the original cut-off grade In summary, the formulation uses transfer functions to
optimisation theory (Lane, 1964). determine the recovered metal content when the net smelter
The estimated annual cash flows or profits are determined return value is computed for each concentrate or product.
by means of a dynamic update of the net smelter return value
that varies over the projected mine life. The dynamic net Marketing constraints
smelter return value is computed for concentrate or product The quantity of liberated and selected ore and gangue minerals
types by applying the defined static and dynamic modifying in concentrates or products are limited by this group of
factors (Liimatainen, 1998; Wellmer, 1989). constraints and based on the market demand or smelters and
refineries short, medium- and long-term sales agreements.
The formulation enables consideration of additive and non-
additive transfer functions that correlate the mineralogical
Smelting constraints
and textural characteristics with liberation and selectivity
This group of constraints restricts the quality of liberated and
processing properties for each geometallurgical domain.
selected ore minerals in concentrates. These constraints also
Geological constraints limit the production capacity of the smelter.
The aim of the geological constraints consists of selecting Refining constraints
orebodies, domains and resource categories in a specific time
The refining constraints group controls the quality of liberated
period of the projected mine life. Consequently, a variety of
and selected ore minerals in concentrates or products. In
scenarios can be analysed with this group of constraints.
addition, this group of constraints limits the production
Mining constraints capacity of the refinery.
The mining constrains simultaneously define the stope geo- Environmental constraints
metry, reserves, mining sequence and production of ore and
The main purpose of this group of constraints consists
gangue minerals over the projected life of mining operations.
limits the quality of liberated and selected gangue or delet-
It is performed by constraining the quantity and quality of ore
erious minerals abundance in waste materials, tailings
and gangue minerals.
and concentrates or products, that in some way have an
The mining constraints define the level intervals, stope environmental impact.
dimensions and locations, cut heights, pillar dimensions and
locations for each orebody. The levels, stopes and cuts can be Financial constraints
also selected in a specific time period to assess any particular This group of constraints controls the mining, processing,
mining scenario. marketing, smelting, refining and environmental fixed
In addition, a number of mining constraints related to the and variable operating costs for each time period over the
underground cut-and-fill mining cycle (Hustrulid and Bullock, projected mine life.
2001) were established in the formulation. Furthermore, mine
capacity, level, stope and cut productivities, mineralogical and APPLICATION
textural characteristics, internal and wall dilution and cut-off
of liberated and selected ore minerals are also considered. A case study was considered to demonstrate the capability of
The stope geometry is then generated by differential cut-off integrating the cut-off of liberated and selected ore minerals
of liberated and selected ore and gangue minerals that vary and the joint cut-and-fill mining and mineral processing
in both horizontally and vertically directions across levels, optimisations. The assessment of the integrated optimisation
stopes and cuts. was performed throughout the definition of several scenarios.
Each scenario included the combinations of static and
The reserves consist of those blocks included in the mining
dynamic modifying factors over the projected mine life.
production of ore and gangue minerals. The optimal solution
defines the quantity and quality of the diluted and blended The geologic setting of the case study consists of several
reserves blocks to be mined and processed for each time mesothermal and pyrometasomatic replacement orebodies in
period and by orebody, domain, level, stope and cut. a thick limestone sequence. The mineralisation is composed
of mainly massive galena and sphalerite with amounts
Processing constraints of chalcopyrite associated with pyrrhotite, arsenopyrite,
The quantity and quality of liberated and selected ore and silicates, sulfates and carbonates.
gangue minerals in concentrates or products is controlled by Two structural and geometric types of orebodies have been
a group of constraints based on mineral processing liberation identified within the project:
and selectivity parameters. 1. a gently dipping sheet, and
The additive and non-additive transfer functions defined 2. steeply plunging chimneys.
for each domain calculate the quantity and quality of the The sheet orebodies comprise a combination of silicates
concentrates or products according to the mineral processing and sulfides, whilst the chimneys are dominated by sulfides
capacity. with or without silicates. One of the irregular sheet orebodies
Normally, a static mineral processing recovery function is was used as the case study, which represents a metasomatic
applied to determine the recovered metal content of specific replacement system. A long-section model of the orebody is
commodity. However, dynamic mineral processing recovery shown in Figure 2.
transfer functions can be applied in the formulation based on A geometallurgical multivariate resource model of
the geometallurgical characteristics and properties of the ore the orebody was previously generated and accessed as
minerals in each domain (Bojcevski, 2003). The expected net fundamental to the optimisation process. The resource
smelter return value could then vary significantly depending model comprised the spatial variability of mineralogical and

THE FIRST AUSIMM INTERNATIONAL GEOMETALLURGY CONFERENCE / BRISBANE, QLD, 5 - 7 SEPTEMBER 2011 265
G TURNER-SAAD

considered in the optimisation process. The processing

selectivity or flotation information was related to the recovery
of ore and gangue minerals. The recovery information was
defined as transfer functions based on the relationship of
the abundance (per cent), association (per cent), particle size
(microns), liberation (per cent) and product size P80 (per
cent) of each ore mineral, in each domain.
The integrated and iterative optimisation process consisted
of assessing the impact of using static and dynamic modifying
factors over the projected mine life, including:
 cut-off and average of liberated and selected ore minerals;
 stope geometries;
 reserves;
FIG 2 - Explicit spatial model of the case study orebody.  ore and waste mining sequences;
 expected mining, processing, smelting and refining
textural characteristics and mineral processing liberation productions; and
parameters. These parameters, in turn, were constrained by  expected profits.
several mutually exclusive geometallurgical liberation and The optimisation was constrained by geological, mining,
selectivity spatial domains within the orebody by applying processing, marketing, smelting, refining, environmental and
and combining implicit modelling and multivariate statistical financial constraints. Four scenarios were considered.
analysis.
Scenario 1:
The spatial models of mineralogical characteristics used
 geological constraints:
in the optimisation process included: abundance (per cent),
association (per cent), particle size (microns) and liberation  static orebody and domains; and
(per cent) of the galena, sphalerite, chalcopyrite, pyrrhotite  static measured and indicated mineral resources;
and arsenopyrite. An example of the galena model is shown  mining constraints:
in Figure 3.  static limits of level intervals;
The mineral processing liberation and comminution  static limits of stope dimensions and locations;
information was also accessed and specifically associated with  static limits of cut heights;
the spatial models of feed size F80 (mm), product size P80  static limits of pillar dimensions and locations;
(microns), throughput (t/h) and energy consumption (kWh/t)  static limits of levels, stopes and cuts dilution rates;
as are shown in Figure 4. The processing recoveries of galena,  static limits of levels, stopes and cuts productivities; and
sphalerite, chalcopyrite, pyrrhotite and arsenopyrite were  static limits of ore and waste production capacities;

Abundance [%] Association [%]

Grain Size [microns] Liberation [%]

FIG 3 - The galena abundance (per cent), association (per cent), particle size (microns) and liberation (per cent) spatial models of the case study orebody.

266 THE FIRST AUSIMM INTERNATIONAL GEOMETALLURGY CONFERENCE / BRISBANE, QLD, 5 - 7 SEPTEMBER 2011
 A CUT-OFF OF LIBERATED AND SELECTED ORE MINERALS OPTIMISATION BASED ON THE GEOMETALLURGY CONCEPT

Feed Size F80 [mm] Product Size P80 [microns]

Throughput [tph] Energy Consumption [kWh/t]

FIG 4 - The feed size F80 (mm), product size P80 (microns), throughput (t/h) and energy consumption (kWh/t) spatial models of the case study orebody.
 processing constraints: limits of the mineral processing liberation or comminution
 static limits of ore production capacity; properties.
 static limits of ore minerals abundance; Scenario 2:
 dynamic limits of liberation or comminution  processing constraints:
properties; and  static limits production capacity of ore,
 dynamic limits of selectivity or flotation properties;  static limits of ore minerals abundance,
 marketing constraints:  static limits of liberation or comminution properties –
 static limits of marketing costs; energy consumption (kWh/t), and
 static limits of shipping costs; and  dynamic limits of selectivity or flotation properties.
 static limits of treatment costs; Scenario 3:
 smelting constraints:  processing constraints:
 static limits of concentrates production capacity; and  static limits production capacity of ore,
 dynamic limits of ore minerals abundance.  static limits of ore minerals abundance,
 refining constraints:  static limits of liberation or comminution properties –
 static limits of products production capacity; and energy consumption (kWh/t),
 dynamic limits of ore minerals abundance.  static limits of liberation or comminution properties –
throughput (t/h), and
 environmental constraints:
 dynamic limits of selectivity or flotation properties.
 static limits of deleterious minerals abundances in
Scenario 4:
waste;
 static limits of deleterious minerals abundances in  processing constraints:
tailings; and  static limits production capacity of ore,
 static limits of deleterious minerals abundances in  static limits of ore minerals abundance,
concentrates or products;  static limits of liberation or comminution properties –
 financial constraints: energy consumption (kWh/t),
 static limits of liberation or comminution properties –
 static limits of mining, processing, smelting and
throughput (t/h),
refining fixed and variable operating costs;
 static limits of liberation or comminution properties –
 static limits of discount rate; and
product size P80 (microns), and
 dynamic limits of metal prices.
 dynamic limits of selectivity or flotation properties.
The geological, mining, marketing, smelting, refining, In summary, the objectives of the four scenarios consisted of
environmental and financial constraints of scenarios 2, 3 assessing the impact of relaxing and constraining the energy
and 4 were similar to scenario 1. The only difference was in the consumption (kWh/t), throughput (t/h) and product size P80
processing constraints and specifically in defining the static (microns).

THE FIRST AUSIMM INTERNATIONAL GEOMETALLURGY CONFERENCE / BRISBANE, QLD, 5 - 7 SEPTEMBER 2011 267
G TURNER-SAAD

RESULTS AND DISCUSSION Liberated and Selected Galena

The probability density function (per cent) and cumulative
distribution function (per cent) of the galena abundance (per
cent) in the resource model is shown in Figure 5. As expected,
both functions were dynamic in the sense that new functions
are computed for every time period. The new functions rep-
resent the statistical distribution of the remaining resources of
the orebody. This means that the quantity of material mined
in a specific time period is removed from the resource model. 0.0 0.5 1.0 1.5 2.0 2.5 3.0
The statistical distribution of the updated resource model Cut-Off [%]
has a direct impact on the definition of the balancing cut-offs
of liberated and selected ore minerals of the following time Ore:Material Ratio Concentrator:Mine Capacities Ratio

periods. The maximisation of the expected economic value

requires mining of high values in each time period as can be Liberated and Selected Galena
seen in both functions of Figure 5.

Liberated and Selected Galena

CDF [%]
PDF [%]

0.0 0.5 1.0 1.5 2.0 2.5 3.0

Cut-Off [%]

0.0 0.5 1.0 1.5 2.0 2.5 3.0 Product:Material Ratio Refinery:Mine Capacities Ratio
Abundance [%]

Probability Density Function Cumulative Distribution Function Liberated and Selected Galena

FIG 5 - The dynamic probability density function (per cent) and cumulative
distribution function (per cent) of the galena abundance (per cent) of each
time period over the projected mine life.

The quantity (t) – quality (per cent) – cut-off of liberated and

selected ore minerals (per cent) plot in Figure 6, also confirm
that the resource model was updated dynamically in each time 0.0 0.5 1.0 1.5 2.0 2.5 3.0
period. This means that the quantity and quality of the galena Cut-Off [%]
is decreasing in every time period and subsequently the cut-
Product:Ore Ratio Refinery:Concentrator Capacities Ratio
off of liberated and selected ore minerals as well.
Liberated and Selected Galena
FIG 7 - The dynamic concentrator-mine (top), refinery-mine (middle) and
refinery-concentrator (bottom) balancing cut-offs of liberated and selected
galena (per cent) of each time period with a static mine, concentrator and
Quality [%]
Quantity [t]

refinery production capacities over the projected mine life.

The behaviour of the information represents the depletion of

the resources in each time period due to mining, processing,
smelting and refining.
0.0 0.5 1.0 1.5 2.0 2.5 3.0
The optimum cut-off of liberated and selected galena (per
Cut-Off [%]
cent) and average of galena abundance (per cent) over the
Material Average of Galena Abundance projected life of the mining operation and by scenario is shown
in Figure 9. The differences in cut-offs among scenarios is due
FIG 6 - The dynamic quantity (t) (material) - quality (per cent) (average of to the limits defined for the energy consumption (kWh/t),
galena abundance) - cut-off of liberated and selected galena (per cent) of throughput (t/h) and product size P80 (microns). The average
each time period over the projected mine life. galena abundance (per cent) decreases over mine life as a
result of the maximisation process.
The dynamic and decreasing balancing cut-offs of liberated Figure 10 describes the material, ore, product productions
and selected ore minerals is illustrated in Figure 7. The (t) and profits ($) of each scenario. The differences among
balancing cut-offs are decreasing due to the updated resource scenarios is due to the static limits defined in some of the
model having higher values. Also, in Figure 7, the static mine, mineral processing liberation properties. However, from these
concentrator and refinery production capacities over the plots can be seen that the refinery capacity is the bottleneck of
projected mine life can be seen. production.
The dynamic mine, concentrator and refinery productions, The energy consumption (kWh/t), throughput (t/h) and
profits and present value plots are illustrated in Figure 8. product size P80 (microns) over the projected life of the

268 THE FIRST AUSIMM INTERNATIONAL GEOMETALLURGY CONFERENCE / BRISBANE, QLD, 5 - 7 SEPTEMBER 2011
 A CUT-OFF OF LIBERATED AND SELECTED ORE MINERALS OPTIMISATION BASED ON THE GEOMETALLURGY CONCEPT

The spatial distribution of the stope geometries and mining

Liberated and Selected Galena
sequences of the four scenarios and the cut-off of liberated
and selected galena are shown respectively in Figures 12 and
13. The spatial differences among the four scenarios is mainly
due to the impact of the defined static mineral processing
liberation or comminution parameters used during the optim-
isation process. Definitively, any change of static or dynamic
parameters involved in the modifying factors has a specific
impact in the stope geometry, reserves, mining sequence, mine
0.0 0.5 1.0 1.5 2.0 2.5 3.0 and processing productions and subsequently the economics.
Cut-Off [%]
CONCLUSIONS
Mine Production Concentrator Production Refinery Production
The main conclusions of this study are:
 the realistic expected maximum economic or net present
Liberated and Selected Galena value of mining operations is reached when the mining
and mineral processing stages are optimised concurrently
instead of isolated;
 the joint mining and mineral processing methodology
enables maximising the depletion of the resources;
 the fundamental information of the optimisation process
is the geometallurgical multivariate resource models,
which integrate the spatial variability of mineralogical and
0.0 0.5 1.0 1.5 2.0 2.5 3.0
textural characteristics and mineral processing liberation
Cut-Off [%]
and selectivity properties;
Mine Profit Concentrator Profit Refinery Profit  the additive and non-additive transfer functions also
need to take into consideration the capability of blending
the ore from mutually exclusive geometallurgical spatial
Liberated and Selected Galena
domains, which contain different mineralogical and
textural characteristics and mineral processing liberation
and selectivity properties;
 the realistic economic assessment of ‘what if’ strategic,
tactical and operational scenarios is obtained when
dynamic modifying factors are applied over the projected
mine life; and
 further research and development is required to enhance
0.0 0.5 1.0 1.5 2.0 2.5 3.0
and apply the methodology to other underground and
Cut-Off [%]
open pit mining methods.
Mine Value Concentrator Value Refinery Value
ACKNOWLEDGEMENTS
FIG 8 - The dynamic mine, concentrator and refinery productions (t) (top), Special thanks to Dr Simon C Dominy for his support and help
profits ($) (middle) and present value ($) (bottom) by cut-off of liberated and in reviewing, commenting on and editing this paper.
selected galena (per cent) for each time period over the projected mine life.
REFERENCES
mining operations is shown in Figure 11. The differences Alford, C, 1995. Optimisation in underground mine design, in
between each scenario is due to the static limits defined for Proceedings APCOM XXV, pp 213-218 (The Australasian Institute
each mineral processing liberation parameter. of Mining and Metallurgy: Melbourne).

Optimum Cut-Off of Liberated and Selected Galena [%] Average of Galena Abundance [%]

0 5 10 15 20 0 5 10 15 20
Time Period [y] Time Period [y]

Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 1 Scenario 2 Scenario 3 Scenario 4

FIG 9 - The dynamic optimum cut-off of liberated and selected galena (per cent) (left) and average of galena abundance (per cent) (right)
over the projected mine life.

THE FIRST AUSIMM INTERNATIONAL GEOMETALLURGY CONFERENCE / BRISBANE, QLD, 5 - 7 SEPTEMBER 2011 269
G TURNER-SAAD

Scenario 1 Scenario 2

0 5 10 15 20 0 5 10 15 20
Time Period [y] Time Period [y]

Material [t] Ore [t] Product [t] Profit [$] Material [t] Ore [t] Product [t] Profit [$]

Scenario 3 Scenario 4

0 5 10 15 20 0 5 10 15 20
Time Period [y] Time Period [y]

Material [t] Ore [t] Product [t] Profit [$] Material [t] Ore [t] Product [t] Profit [$]

FIG 10 - The dynamic material (t), ore (t), concentrate or product (t) productions and profits ($) over the projected mine life.

Energy Consumption of the Liberation Process [kWh/t] Throughput of the Liberation Process [tph]

0 5 10 15 20 0 5 10 15 20
Time Period [y] Time Period [y]

Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 1 Scenario 2 Scenario 3 Scenario 4

Product Size P80 of the Liberation Process [microns]

0 5 10 15 20
Time Period [y]

Scenario 1 Scenario 2 Scenario 3 Scenario 4

FIG 11 - The energy consumption (kWh/t) (top), throughput (t/h) (middle) and product size P80 (microns) (bottom) of the liberation process over the projected mine life.

270 THE FIRST AUSIMM INTERNATIONAL GEOMETALLURGY CONFERENCE / BRISBANE, QLD, 5 - 7 SEPTEMBER 2011
 A CUT-OFF OF LIBERATED AND SELECTED ORE MINERALS OPTIMISATION BASED ON THE GEOMETALLURGY CONCEPT

Scenario 1 Scenario 2

Scenario 3 Scenario 4

FIG 12 - The stope geometries and mining sequences of a cut-and-fill mining over the projected mine life.

Scenario 1 Scenario 2

Scenario 3 Scenario 4

FIG 13 - The spatial distribution of the cut-off of liberated and selected galena (per cent) of a cut-and-fill mining over the projected mine life.

THE FIRST AUSIMM INTERNATIONAL GEOMETALLURGY CONFERENCE / BRISBANE, QLD, 5 - 7 SEPTEMBER 2011 271
G TURNER-SAAD

Asad, M W A, 2005. Cut-off grade optimization algorithm for open Napier, J A L, 1983. The effect of cost and price fluctuations on the
pit mining operations with consideration of dynamic metal price optimum choice of mine cutoff grades, Journal of the Southern
and cost escalation during mine life, in Proceedings APCOM African Institute of Mining and Metallurgy, 83(6):117-125.
2005, Tucson, Arizona, pp 273-277 (A A Balkema Publishers). Ovanic, J and Young, D S, 1995. Economic optimization of open
Ataee-pour, M and Baafi, E Y, 1998. Implementation of a heuristic stope geometry using separable programming with special
algorithm to optimise stope limits with excel modules, in branch and bound techniques, in Proceedings Third Canadian
Proceedings APCOM 1998, Kalgoorlie, pp 161-164. Conference on Computer Applications in the Mineral Industry
Baird, B K and Satchwell, P C, 2001. Application of economic (ed: K Dagdelen), pp 129-135.
parameters and cutoffs during and after pit optimization, Mining Rahal, D, Smith, M L, Van Hout, G and Von Johannides, A, 2003.
Engineering, 53(2):33-40. The use of mixed integer linear programming for long-term
Bojcevski, D, 2003. Metallurgical Characterisation of George scheduling in block caving mines, in Proceedings Application of
Fisher Mesotextures and Microtextures, 369 p (The University of Computers and Operation Research in the Minerals Industries,
Queensland: Brisbane). pp 1-9 (Southern African Institute of Mining and Metallurgy:
Marshalltown).
Dagdelen, K, 1993. An NPV maximization algorithm for open
pit mine design, in Proceedings APCOM XXIV Application of Rudenno, V, 1979. Determination of optimum cutoff grades, in
Computers and Operations Research in the Mineral Industry, Proceedings 16th Application of Computers and Operations
pp 257-263 (Canadian Institute of Mining, Metallurgy and Research in the Mineral Industry, pp 261-268 (Society of Mining
Petroleum: Montreal). Engineers of the American Institute of Mining, Metallurgical, and
Petroleum Engineers, Inc).
Davis, G A and Newman, A M, 2008. Modern strategic mine
planning, in Proceedings 2008 Australian Mining Technology Smith, M L and O’Rourke, A, 2005. The connection between
Conference, Sunshine Coast, Queensland, pp 1-13. production schedule and cut-off optimization in underground
mines, in Proceedings 32nd Application of Computers and
Fourer, R, Gay, D M and Kernigham, B W, 1993. AMP A Modelling Operations Research in the Mineral Industry (eds: S Dessureault
Language for Mathematical Programming, 351 p (Boyd & et al) (Society for Mining, Metallurgy, and Exploration: Tucson).
Fraser Publishing Company, International Thomson Publishing:
Danvers). Smith, M L, Sheppard, I and Karunatillake, G, 2003. Using MIP
for strategic life-of-mine planning of the lead/zinc stream at
Grieco, N, 2004. Risk analysis of optimal stope design: Incorporating Mount Isa Mine, in Proceedings Application of Computers and
grade uncertainty, MPhil thesis, The University of Queensland, Operation Research in the Minerals Industries, pp 1-10 (Southern
Brisbane. African Institute of Mining and Metallurgy: Marshalltown).
Hustrulid, W A and Bullock, R L, 2001. Underground Mining Thomas, G and Earl, A, 1999. The application of second-generation
Methods: Engineering Fundamentals and International Case stope optimisation tools in underground cut-off grade analysis,
Studies, 718 p (Society for Mining, Metallurgy and Exploration in Proceedings Strategic Mine Planning Conference, pp 1-6 (The
Inc: Littleton)p. Australasian Institute of Mining and Metallurgy: Melbourne).
IBM 2011. IBM ILOG CPLEX Optimizer [online]. Available from: Topal, E, Kuchta, M and Newman, A, 2003. Extensions to an
<http://www-01.ibm.com/software/integration/optimization/ efficient optimization model for long-term production planning
cplex-optimizer/> [Accessed: 6 May 2011]. at LKAB’s Kiruna mine, in Proceedings Application of Computers
JORC, 2004. The JORC Code, Australasian Code for Reporting of and Operation Research in the Minerals Industries, pp 289-
Exploration Results, Mineral Resources and Ore Reserves, 31 p 293 (Southern African Institute of Mining and Metallurgy:
(The Joint Ore Reserves Committee of the Australasian Institute Marshalltown).
of Mining and Metallurgy, Australian Institute of Geoscientists Trout, L P and Grice, A G, 1993. Optimisation of underground mine
and Minerals Council of Australia). activity scheduling, in Proceedings Australian Conference on the
Lane, K F, 1964. Choosing the optimum cut-off grade, Quarterly of Application of Computers in the Mineral Industry (ed: E Y Baafi),
the Colorado School of Mines, 59(4):811-829. pp 310-315 (University of Wollongong: Wollongong).
Lane, K F, 1979. Commercial aspects of choosing cutoff grades, in Turner-Saad, G, 2010. Vision for a risk adverse integrated
Proceedings 16th Application of Computers and Operations geometallurgical framework, in Proceedings 42nd Annual
Research in the Mineral Industry, pp 280-285 (Society of Mining Meeting of the Canadian Mineral Processors, Ottawa, pp 197-
Engineers of the American Institute of Mining, Metallurgical, and 213.
Petroleum Engineers, Inc). Turner-Saad, G, 2011. A joint cut and fill mining and mineral
Lane, K F, 1988. The Economic Definition of Ore, 147 p (Mining processing methodology for the strategic mine planning process,
Journal Books: London). in Proceedings Second International Seminar on Mine Planning,
Antofagasta, pp 76.
Lane, K F, Hamilton, D J et al, 1984. Cutoff grades for two minerals,
in Proceedings Application of Computers and Mathematics in Turner-Saad, G and Smith, M L, 2006. The impact of the bulk
the Mineral Industries, pp 485-491 (The Institution of Mining density and metallurgical recovery in strategic cut and fill mining,
and Metallurgy: London). in Proceedings JKMRC International Student Conference II,
pp 187-200 (The University of Queensland: Brisbane).
Liimatainen, J, 1998. Valuation model and equivalence factors
for base metal ores, in Proceedings Seventh International Van Leuven, M A, 1998. Risk analysis-an aid in selecting an
Symposium on Mine Planning and Equipment Selection (ed: R K underground mining method, in Seventh International
Singhal), pp 317-322 (A A Balkema: Calgary). Symposium on Mine Planning and Equipment Selection, pp 349-
354 (ed: R K Singhal) (A A Balkema: Calgary).
Murray, R M and Magri, E J, 1978. The Use of Linear Programming
in the Short-Term Planning of Stoping Production in Gold Wellmer, F W, 1989. Economic Evaluation in Exploration, 163 p
Mines, pp 262-268 (Southern African Institute of Mining and (Springer-Verlag: Berlin).
Metallurgy: Marshalltown).

272 THE FIRST AUSIMM INTERNATIONAL GEOMETALLURGY CONFERENCE / BRISBANE, QLD, 5 - 7 SEPTEMBER 2011