Strategic Open Pit Planning and Optimisation

STRATEGIC OPEN PIT
PLANNING AND OPTIMISATION
Level 1
Presented by Richard Gawthorpe

February 2005
Strategic Open Pit Planning and Optimisation - Level 1 1
TABLE OF CONTENTS
1. Introduction ....................................................................................................................... 2
2. Lerch-Grossmann Background ......................................................................................... 4
3. Economic Parameters Preparation .................................................................................. 5
4. Block Model Preparation................................................................................................... 8
5. Block Model Import to NPVS .......................................................................................... 10
6. Economic Model Generation........................................................................................... 11
7. Cut-Off Grades and Block Allocation .............................................................................. 12
8. Mining Factors ................................................................................................................ 13
9. Cost Adjustment Factors................................................................................................. 14
10. Block Value Calculations ............................................................................................ 15
11. Ultimate Pit Generation .............................................................................................. 16
12. Slope Regions and Pit Limits...................................................................................... 19
13. NPV Calculation.......................................................................................................... 20
14. Ultimate Pit Results .................................................................................................... 21
15. Graph Output .............................................................................................................. 22
16. Reserve Reporting ...................................................................................................... 24
17. Plans And Sections .................................................................................................... 25
18. Case Study Management ........................................................................................... 26
19. Ultimate Pit Analysis ................................................................................................... 27
20. Existing Pit Designs .................................................................................................... 28
21. Sensitivity Studies....................................................................................................... 29
22. Pit Surface Export ....................................................................................................... 30
23. Optimisation and Practical Open Pits ......................................................................... 31
24. Pushback Generation ................................................................................................. 32
25. Pushback Customisation ............................................................................................ 34
26. Project Overview......................................................................................................... 35
27. Terminology ................................................................................................................ 36
February 2005
1. Introduction
These notes are intended to accompany a course given in strategic open pit planning and
optimisation. The principal piece of software connected with this course is the NPV
Scheduler (Version 3.2). The primary emphasis of this Level I course is to train mining
engineers and geologists up to the stages of generating an ultimate open pit and associated
pushback expansions. A corresponding Level II course deals with production scheduling and
other more detailed aspects of open pit optimisation.
These notes are intended to complement the full and extensive documentation provided with
the NPV Scheduler, which deal with much more detail regarding the software operation.
An overview of the NPV Scheduler system is shown in the flowchart overleaf.
The main components of the NPV Scheduler can be summarised as:
Main Components
Model Import
Economic settings
Ultimate pit
Pushback generation
Scheduling
Stockpiling
The overall preparation stages for the work with the NPV Scheduler will normally include the
following elements.
Preparation
Block model
Economic parameters
Processing parameters
Operating parameters
Slope parameters
Pit limits
Topographical surfaces
Design surfaces
Scheduling requirements
Stockpile data
February 2005
NPV Scheduler - Overview
Import Geological Block Model
Metal prices/ Mining and

Economic Calculations
selling costs processing costs
Economic Model:
Slope regions
VALUE
Slope Settings Pit Shell Optimisation Pit Limits
Pit Result Model:

Reports/graphs
PHASE
for analysis
SEQUENCE
Select Particular Ultimate

Pit
Tonnage/size
Pushback Generation Physical Limits
parameters
Reports/graphs
Pushback Model
for analysis
Pushback
Scheduling Targets
Controls
Cashflows/ Scheduled Block

Reports/graphs
production data Model
Key
Process
Data
February 2005
2. Lerch-Grossmann Background
Since the 1960s, computerised open pit optimisation methods have been evolving. Now,
most major mining companies use some form of these methods in the initial and ongoing
design of their open pit operations. They can also be used to assist in the transition from
open pit to underground mining. Examples of widely used systems include the Four-X pit
optimisation software and the NPV Scheduler. The primary focus of these software systems
is to assist in the determination of the optimum size and shape of open pits, as well as to
enable the generation of production schedules. The systems encompass a number of
processes which utilise computerised geological block models and allows detailed economic
analysis of different open pit options. This analysis, which includes discounted cashflows,
also yields extremely valuable information to then assist in the mine planning and scheduling
of open pits.
Lerchs and Grossmann publish (1964) mathematical method for

optimal pit shape.
Three-D (1985) implementation on a range of computers. Single pit

with a maximum value.
Other optimisation methods
Four-X, NPV Scheduler

- optimal shells
- discounted cash flows
- maximising of NPVs and other criterion
Lerchs and Grossmann published a mathematical method, in 1964, for the determination of
the optimal shape for an open pit in three dimensions. This used an orebody block model,
and generated lists of related blocks that should, or should not, be mined. The final list
described the pit with the highest possible total value, subject to the required pit slopes.
The Three-D program, released by Whittle in 1985, was a computer implementation of the
Lerchs-Grossmann method. This used a regular block model, whose blocks have economic
values that represent the net cash flow that would result from mining the block in isolation.
Optimal pit outlines produced did not use discounted cash flows.
The Floating Cone method, as the name implies, repeatedly searched for and checked the
total value of block groups forming inverted cones. Whole cones were assigned for mining if
their total value was positive. This procedure continued until no more positive cones were
found. However, the solutions produced by this method could not guarantee the final pit was
optimum. Other groups also implemented a two-and-a-half dimensional Lerchs-Grossmann
algorithm, which used the supplied slopes only in the north, south, east and west directions.
The Four-D (and subsequently Four-X) and NPV Scheduler programs use the same Lerchs-
Grossmann technique to produce a set of nested optimal pits. Each pit is optimal for a
different set of value ratios, and these are used to guide different mining schedules. Financial
analysis of these schedules, with discounted cash flows, allows selection and sensitivity
analysis of the best pit according to different criteria.
February 2005
3. Economic Parameters Preparation
Develop spreadsheet model
Include all costs that stop when mining stops
Sometimes useful to think in terms of contractors
Time and replacement capital costs

- Factored into cost/tonne for optimisation
Extra ore mining costs must be added to processing costs
The normal and most convenient method of preparing economic data is to first prepare a
spreadsheet model, for input and appropriate processing of relevant cost information. This
will also normally be accompanied by processing data, as some of the processing
throughputs are necessary for cost manipulation and other processing-related values may be
required for open pit analysis. The spreadsheet can be customised according to the
particular project requirements, producing final quantities that are suitable for direct use as
optimisation parameters.
The most important premise is to include in some way all costs that would stop if the mine
stopped production. For example, certain sunk capital costs will normally be excluded by this
consideration. Another useful guide in this debate is to sometimes consider how pit costs
would be considered if contractors were being used. They would normally submit a
cost/tonne figure that would include a number of operational and capital costs, but this overall
cost/tonne figure would normally be directly used for pit optimisation.
Time (e.g. general and administration) and replacement costs will not normally be available in
a cost/tonne form. It is, however, necessary to determine component cost per tonne figures
that can be accrued for pit optimisation purposes. In order to convert such costs, the type of
production limitation at the mine must first be chosen - the commonest form is probably mill-
limited. Based on the relevant limitation type, the time cost is divided by the corresponding
production capacity, to yield an effective cost/tonne figure.
Frequently, the cost per tonne of mining ore will be greater than the cost of mining waste, e.g.
due to different equipment. In this case, the additional (i.e. the difference) cost incurred must
be calculated and then added onto the processing cost. Therefore the processing cost for
optimisation purposes actually includes cost components for the mining of ore as well pure
processing.
February 2005
EXAMPLE - PIT OPTIMISATION ECONOMIC PARAMETERS
CATEGORY PARAMETER UNIT VALUE

Gold Price $/g 9.645
Selling Cost $/g 0
Processing
Cost
Oxide $/t 7.35
Transition $/t 7.35
Sulphide $/t 9.65
Recovery
Oxide 95.3%
Transition 95.3%
Sulphide 93.5%
Mining Costs
Reference cost $/t 1
Incremental cost
Ore $/t/bench 0.01
Waste $/t/bench 0.01
Oxide
Ore $/t 1.04
Waste $/t 0.95
MCAF 0.95
Additional ore mining $/t 0.09
Transition
Ore $/t 0.97
Waste $/t 0.89
MCAF 0.89
Sulphide
Ore $/t 1.08
Waste $/t 0.98
MCAF 0.98
Total Ore Costs
Oxide $/t 7.44
Transition $/t 7.43
Sulphide $/t 9.75
Marginal
Cut-Off Grades Oxide g/t 0.81
Transition g/t 0.81
Sulphide g/t 1.08
Key
Bold Supplied
Italic Derived
Used directly in optimisation
February 2005
Economic Parameters' Preparation

Main Components
Mining costs
Processing costs
Prices
Selling costs
Discount rates
Profit' model
Mining Costs' Preparation

$/t or $/m3
Costs initially calculated for reference block
Mining cost adjustments
By rock type
By $/tonne/bench for each process method
Rehabilitation costs
Care with capital costs
Processing Costs and Parameters

$/tonne of ore by rock type and processing method
Additional $/unit of metal output by rock type and process
Recovery by rock type and processing method
May be a function of input grade
Need to build in time costs
Cut-Off Grades
Good idea to check-calculate in spreadsheet
NPV can determine it's own COGs
Can impose COGs
For multi-elements, options include:
Overall revenue
Any element being above its cut-off
All element being above their cut-offs
Other Operating Parameters

Ore throughput
Dilution
Recovery
Pushback control:
Access space requirements
Minimum size
Tonnage requirements
Depth/physical limits
Block Model Preparation

Important Considerations
Coverage
Block size
Rock types
Grades
Density/tonnage
Cost adjustments
Pit limits
February 2005
4. Block Model Preparation
Tonnage direct or by calculation with DENSITY field.
Grades - % or g/t
Rock type single numeric field describing all processing rock types.
As only one numeric rock field can be input and utilised in optimisation, some careful planning
is important to consider all the different what-if? type processing scenarios that may be
required, what reporting is required. The rock field used for optimisation may well be
combination of, say, resource class and geological rock type.
It is important that sufficient waste blocks outside of the potential ore bodies are available for
the optimiser to consider when generating potential pits. This may well required modification
of the block model used for purely resource estimation purposes.
The NPVS software can directly read in sub-cells, although the individual positions of the sub-
cell contents will simply allocated to the parent cell. Regularised models may also be
supplied.
Costs initially calculated for REFERENCE block
Mining cost adjustments can be supplied as:

1. Defined per rock type in NPVS
2. As incremental adjustments $/tonne/bench by ore type
3. As pre-defined MCAF values for each block
Processing cost adjustments can be supplied as:

1. Through supplied formulae (on grades) by rock type.
3. As pre-defined PCAF values for each block
Rock type mining CAF = Mining cost for rock type

Reference waste mining cost
All costs are initially calculated, and then supplied to the optimisation software, for some
reference block position. This block does not necessarily have to exist; it is purely used as a
point of reference. Cost adjustment ratios are then used for all other blocks. Such ratios can
be used for mining and/or processing costs. If multiple cost adjustment are supplied, e.g. by
rock type and by a defined MCAF in the block model, then in the block value calculation both
adjustment factors will be multiplied together.
It is common practice to often also use the economic data preparation spreadsheet for the
calculation of any necessary cost adjustment factors. This enables the different cost
components to be combined as required, with the necessary addition of time costs, as well as
the ratio calculation against the reference block.
The Check Value utility can be used to display the block value. This is a very useful method
of checking that the cost adjustment factors have been supplied and are being used in the
correct way.
February 2005
Project Start-Up
Working directory
Project name
Initial CASE structure
Model Import
Model name (e.g. model.dm)
Required fields e.g.:
DENSITY
ROCKCODE
AU
Choose data type for each field, e.g.
DENSITY Density
ROCKCODE Rock type
AU Product
Product unit:
Grade
Mass
Percentage
Unit confirmation
Potential Data Types

Attribute
Density
Product
Mining CAF
Processing CAF
Ore Percent
Ore Fraction
Rock Type
Mining Surface
Property Surface
Tonnage
Volume
Input Model Report

Model statistics
Field assignments
Units
Parameter Labels
Reporting
Descriptive
February 2005
5. Block Model Import to NPVS
Grades or metal quantities can be supplied.
Tonnages can be supplied directly or calculated.
Rock type field (numeric)
Mining and/or processing cost adjustment factors (optional)
Other controlling surface or attribute fields (optional)
When the block model is imported into NPVS, the required fields must be identified. After the
type of identified must be selected.
Products can be supplied as grades, from which metal quantities will be calculated.
Alternatively metal mass quantities can be defined. These quantities can also represent pre-
calculated economic values. For each product entered as a grade, the user also select
whether it is a metal unit/tonne type field (e.g. g/t) or whether it is percentage.
Tonnages can be calculated from the supplied cell (and sub-cell) sizes and the supplied
density values. Alternatively tonnages may be supplied directly. Additional ore fraction or ore
percent fields may also be defined. These may be useful for indicator-type models or
regularised model, where the contained tonnages are in actual fact a proportion of the
physical blocks they occupy. In all cases, a formula is shown describing exactly how the
block tonnages are going to be calculated.
A default density is also prompted for. This is normally entered as zero, meaning that any
parts of the model prototype that do not contain blocks can be assumed as air.
The supplied rock type field must be numeric. A list is built up of all the unique rock type field
values found in the imported block model. Customised parameters labels (e.g. RockType 0 =
Waste) may also be defined to simplify later reporting.
After model import, the report of model contents should be inspected.
Prior to model import, it always good practice to evaluate the model in detail, by the
optimisation rock type, and based on supplied cut-offs. These results can then be used for
subsequently validation purposes.
Block Model Rock Types

Include all details that might eventually be required
Should not re-import if extensive rock type changes
Resource considerations
Processing/reporting
February 2005
6. Economic Model Generation
Cost calculations required?
Data for reporting
Prices
Mining costs, factors and adjustments
Processing parameters
All of the supplied parameters here control how economic values are determined in each
block of the imported model. The imported model may already contain pre-calculated
economic values, and these may be used directly. However, it is more normal to use the
NPVS system with supplied parameters to calculate economic values.
Data fields for reporting must be defined. This step is important as later reports can become
unnecessarily cluttered if too many fields are requested. The number of processing methods
needs to be considered, as well as the number of rock types that may be processed.
Prices and selling costs are entered for each of the different products (metals). The user
must also select what type of cut-off test is required. This is covered in much more detail on
the section 7.
The default mining cost may be entered as a cost per tonne or per unit volume. Mining
factors may also be entered these are dealt with in much more detail in section 8.
Processing parameters are split for each rock type. A separate mining cost adjustment factor
and rehabilitation cost may be entered for each rock type. A number of processing methods
may also be defined for each rock type.
For each processing method, the following parameters are required:
- Processing cost ($/tonne feed), which may be a formula with respect to grades
- Treatment recovery (may be a formula) for each metal
- Additional processing cost per unit of product, for each metal (optional)
- Defined cut-offs (optional), for each metal
The waste rock type mining CAFs are also applied in the mining cost calculations.
Additional mining cost adjustment, $/tonne/bench, may also be defined for waste and the ore
types associated with each processing method. For this option, the mining costs can be
increased both below and above the selected reference bench.
With all the parameters defined, the economic model may be generated. This calculates for
every block:
- Total mining cost

- Total processing cost (if potential ore present)
- Total revenue (if potential ore present)
- Net economic value
These quantities can be colour-coded in plan or sectional views through the block model. A
summary report is also produced, evaluating the model, and clearly listing all of the input
parameters.
February 2005
7. Cut-Off Grades and Block Allocation

Of fundamental importance in the block value calculation is the basic decision as to whether
mineralised parcels within each block are going to be sent to processing or not. If they are
not processed, then these parcels contents will be treated as waste. If they are processed,
the processing costs and subsequent revenues need to be accounted for in determining the
blocks net economic value. There are 4 different ways in which NPVS can apply the process-
reject test.
Revenue > rehabilitation cost
Sum of (grades / cut-offs) > 1
All grades > cut-offs
Any grade > cut-off
The first (and default) method is to consider whether the potential revenue-processing cost
exceeds the rehabilitation cost (for waste). This can be thought of from the respect of a truck
arriving at the edge of the pit. If more money can be made by processing the trucks contents
than by dumping the material as waste, the truck will be sent to the mill. If the model has only
one element and a zero rehabilitation cost, then this test is the same as applying the marginal
cut-off grade determined from the NPVS parameters. This calculated marginal cut-off grade
is always calculated and reported. This test method is the best method in terms of producing
the highest economic values possible, both for optimal pits and on a block-by-block basis. All
of the other test methods can only produce lower economic values.
The calculated cut-off grade is determined by the formula below:
Cut-Off = Processing Cost x (1+Dilution)

-------------------------------------------------------------------
(PriceSelling CostAdditional Treatment Cost)
If any of the other test methods are selected, then it is possible to impose supplied cut-off
grades for each rock type/processing method/metal combination. If a minimum cut-off is not
supplied, the calculated cut-off will be used. For the second method, the grade/cut-off ratios
are added up, and if the total is greater than 1, the block is treated as ore.
rd
For the 3 method, all of the available grades have to be greater than the corresponding cut-
offs. For the 4th method, only one of the available grade fields has to be greater than the
corresponding cut-offs.
Reasons for defining cut-offs may include:
- Because the block model already dilution built into it, at a particular cut-off grade.
- Because grade control is applied at a particular cut-off grade.
- Requiring one particular metal to dominate ore/waste selection (method 4)
- Experimenting with the effect of cut-off on pit optimisation.
- Comparison with other evaluations at particular cut-off grades.
February 2005
8. Mining Factors
There are basically two ways in which mining factors can be applied to pit optimisation. The
block model can be prepared beforehand, so that blocks already have dilution embedded into
them. If this method is used, then cut-off grades normally must have been applied, to ensure
the engineered ore blocks are selected properly.
Alternatively, global dilution and recovery factors can be applied as parameters directly in
NPVS. The principal mining factors which can be applied directly in NPVS are dilution and
mining recovery. There are advantages and disadvantages associated with both methods.
Pre-prepared Mining Factors Advantages
- Can be engineered in great detail,

- Can account for variations by location, rock types.
- Directly reports accurate ore quantities as supplied to processing.
Pre-prepared Mining Factors Disadvantages
- May be more difficult to test different cut-off grades and economic scenarios.
- May be more difficult to plot original in-situ grades on plans and sections.
NPVS Mining Factors Advantages
- Easy to apply and vary for different scenarios.

- Easy to test different cut-off grades and economic parameters.
NPVS Mining Factors Disadvantages
- Factors only applied globally.

- Only applies zero dilutant grades.
- Ore quantities are reported without dilution.
When the NPVS dilution is applied, the following effects include:
- Unit processing costs are increased

- The economic cut-off grade is increased.
- The treatment recovery will be affected if it is a function of grade.
- Values of ore blocks will be reduced.
- Ore tonnages and grades will not be reported as diluted (still reported as in-situ).
There may, however, be less ore than undiluted, purely because of the cut-off and
recovery changes.
- Recovered products may be affected because of less ore.
When the NPVS mining recovery is applied, the following effects include:
- Ore block values are reduced because less ore is processed. Cut-off grades are not
affected.
- Recovered products decrease.
- Reported in-situ ore tonnages and grades are not affected (still in-situ).
February 2005
9. Cost Adjustment Factors

Adjustments to mining costs can be applied in a number of ways:
Mining cost adjustment factors by model block
Mining cost adjustment factor by rock type
Mining cost adjustment factor $/tonne/bench by process type
For mining cost adjustment factors supplied with the imported block model, values will have
had to be generated previously, normally using the mining software system involved in
creation of the block model. This allows almost any type of variation in mining cost with
position and/or rocktype to be accommodated. These values are ratios, and if greater than 1,
indicate a higher cost than the supplied economic parameters. The cost for each block then
is determined by multiplying the cost adjustment factor in any particular block with the
supplied cost/tonne.
Overall mining cost adjustment factors can also be supplied by rock type. This is also simply
a direct numeric ratio, to be multiplied by the default mining cost. This facility allows different
costs to be tested very quickly within various NPVS case studies.
An additional mining cost adjustment, in terms of $/tonne/bench can be applied by process

type. This allows depth effects to be modelled economically. It is split by process type, to
allow for different process (or waste) destinations.
If more than one type of mining cost adjustment is applied, they are simply multiplied together
when the block values are calculated.
Processing cost adjustments can be applied in two ways:
Processing cost adjustment factors by model block
Processing cost formulae
As with mining costs, a processing cost adjustment factor can be supplied with the imported
block model. In this case values will have had to be generated previously, normally using the
mining software system involved in creation of the block model. This allows almost any type
of variation in processing cost with position and/or rocktype to be accommodated.
Instead of entering the processing cost/tonne directly, a formula can be entered. This formula
is a function of the available grade fields in the input model.
February 2005
10. Block Value Calculations

Assume block is already uncovered
Assume block will be mined
Any cost that will stop if mining stops must be included
Any cost that continues if mining stops must be excluded
No allowance is made for the cost of stripping, during the block value calculation, because
that is precisely what the optimizer calculates. Similarly, any cut-off grades applied during
optimisation do not reflect the cost of stripping.
Values are calculated on the assumption that the block will be mined. The optimizer may well
decide not to mine a block whose net value is negative. However, if the block has to be
mined to access another more valuable block, any ore in the original block will be mined to
help pay for stripping, as would happen in practice.
Not all costs associated with mining are input directly into pit optimisation, because costs that
continue, even when the mine may stop, do not have any direct effect on the size of the pit
which will yield the highest economic benefit. A typical example is the sunk (unsalvageable)
capital cost for the processing mill. Although these costs need to be ultimately subtracted to
determine the overall net present value of a project, they should not be used as costs in the
optimisation itself.
The block calculation process in NPVS may be summarised as follows:
1. Potential Revenue (PR) = (Price-Selling Cost)*Recovery*Grade*Mining Recovery

2. Ore Cost (OC) = Ore Costs*Mining Dilution*PCAF
3. Product Cost (PC) = Unit Cost*Recovery*Grade*Mining Recovery
4. Determine Economic Cut-Off = OC/(PR+PC)
5. Determine ore or waste, according to selected method
6. Mining Cost (MC)= Reference Cost* MCAF*rock type adj*bench adj
7. If ore: Net VALUE = PR OC PC MC
8. If waste: Net VALUE = -MC
The NPVS system also has a Check Value facility, where all of the block value calculations
involved with the supplied NPVS economic parameters are shown in an interactive grid
(spreadsheet). Any block grades and tonnages may be entered, and the corresponding
calculations updated. This is an extremely powerful validation tool, and helps greatly in
understanding of how a block values are calculated.
February 2005
11. Ultimate Pit Generation
Slopes
Phase generation
NPV calculation
Pit limits
Based on the generated economic block model, and a supplied set of overall slope angles,
NPVS will determine an ultimate pit that yields the highest possible cashflow (maximum
profit). Detailed control of slope angles is discussed in Section 12.
After attaining the maximum cashflow pit, a number of (normally) internal pit phases (shells)
are also determined. Each of these phases represents an alternative maximum cashflow pit,
in which one of the economic parameters has been varied from the supplied parameters. A
whole series of phases can be produced, each of which represents a different factor which
has been applied to one of the following variables:
- Profit
- Price
- Mining cost
-
These internal phases are important for the selection of alternative optimal pits, for the
generation of mining sequences and for the demarcation of pushback expansions.
For all of the blocks inside the ultimate pit, an idealised optimal extraction sequence (OES) is
also generated. This OES strives to attain the highest discounted cashflow possible, based
on the supplied discount rate and ore processing rate. The phase structure created for a
given Ultimate Pit is used as the basis for constructing the OES (a block by block extraction
sequence), however the OES is selected to produce the highest discounted cashflow,
whereas each phase has been constructed to represent the highest undiscounted cashflow
for given economic parameters.
It may occur that some blocks towards the end of the ultimate pit may not add onto the DCF,
in which case a smaller pit than the ultimate pit may yield the highest DCF. However, it is up
to the user whether all blocks from the ultimate pit should still be included.
A maximum revenue factor of more than 100% can be defined, in which case the largest
ultimate pit produced will be a pit past the peak, in terms of the base economic parameters.
Other control options for the optimisation of the ultimate pit include:
- Additional discounting for of bench contents by depth

- Maximisation of resources contained in the ultimate pit
- Use of an existing pit shell optimisation just generates internal phases
February 2005
ULTIMATE PIT CONTROL

Overall Controls
Pit limits
Slope regions
Settings
Pit Limits
Import/define strings
Copy trace > pit limit
Define type of limit:
Global rectangular bounds
Local limits - cannot encroach
Data information - check
Save and exit
Can enable/disable
Slope regions
Import/define strings
Copy trace > pit limit
Annotate to check serial numbers
Can enable/disable
Separating surface file
Separating surfaces
Import surfaces individually
Combine with:
Edit Surface > Append
Settings
Copy case settings
Ultimate pit
Sequencing
Slopes
Time
Advanced
February 2005
ULTIMATE PIT SETTINGS

Ultimate Pit Settings
Maximise cash flow (Lerchs-Grossmann)
Underground option?
Top-down discounting?
Maximise resources
Use existing pit shell:
Surface file
Shell definition
Sequencing Settings
Optimize NPV
Phase factors - profit, price or mining cost
Profit, price or mining cost factors
Max. revenue factor
Incremental factor
Include all blocks?
Optimize blending
Targets
Slopes' Settings
Split horizontally by regions
Split vertically by either:
Bench Level
Blocks beneath topography
Supplied surfaces
Slope definitions:
Filter dimensions
Azimuth/slope angle pairings
Time Settings
Discount rate
Average ore output rate
Advanced Settings
Reblocking:
X/Y/Z
Ultimate Pit
Phases
Initial topography
No. of points for charting extraction sequence
February 2005
12. Slope Regions and Pit Limits

A slope region is a physical volume, to which a particular group of overall slope angles and
corresponding azimuths are defined. Slope angles in between defined azimuths are
interpolated. Different slope regions can be set up by:
Horizontal bench levels (above/below supplied reference)
Above/below a surface n blocks below supplied topography
Split horizontally, according to supplied slope region perimeters
Split vertically, according to supplied separating surfaces
If no controlling physical data (perimeters or surfaces), then only the first two methods can be
used. If either perimeters or surfaces are supplied, then the slope regions are demarcated by
their XY partition or Section Index.
Slope region perimeters are set up in an interactive slope region editor. In this process,
regions can be defined interactively, or they can be created by importing existing perimeters
from other sources. When defining the angles associated with each slope region, it will be
found that there is always one more slope region (XY partition) than there are perimeters.
This extra slope region is simply the remainder of any blocks that have not been included
within any of the defined slope region perimeters. The same logic pertains to separating
surfaces.
When surfaces are supplied, Section Index 1 starts at the top. Multiple surfaces, defining
multiple sections for slope regions, can be set up by using the Edit Surface File facility. This
allows the combination of various individual surfaces, which may have been imported or set
up automatically when importing the block model.
When the Ultimate Pit is generated (i.e. the optimisation is run), the results gives a full list of
all the slope regions and their constituent angles and azimuths. There is also a Check Slope
facility which indicates how closely the required slopes can be modelled, based on the block
sizes and the slope filters (number of blocks in X and Y) being during slope projection.
Two types of Pit Limits can be defined:
Local pit restrictions (can be complex, with associated costs and limiting depths)
Global rectangular bounds
Local pit restrictions can be any polygon, within which ultimate pit creation will be prevented.
Specific controlling depths can also be applied to these restrictions. Alternatively, costs
associated with the moving of these restrictions (i.e. to mine within them), which can then be
accounted for during optimisation. Such restrictions might typically be used to represent plant
infrastructure, buildings, rivers or roads.
Local pit restrictions, with appropriate limiting depths, may also be used to prevent
excessively small drop-cuts (inverted pyramids) forming during optimisation.
A global rectangular bound is an overall X-Y set of limits, outside of which no ultimate pit can
be created. Even if a complicated polygon is defined, it will be turned into a rectangular
bound, based on the maximum X-Y limits of the polygon.
February 2005
13. NPV Calculation
DCF calculation based on supplied ore processing rate
Optimum extraction sequence (OES) strives for highest NPV
Real NPV is re-calculated during subsequent scheduling
The supplied discount rate and ore processing rate values are used to determine the
discounted cashflow. The sequence used for allocation of blocks with respect to time is
determined so as to yield the highest total discounted cashflow. In most cases, this best
sequence and associated cashflows will be simplistic, as actual schedules will require many
more controls and criteria applied. However, this optimum extraction sequence (OES), is still
very useful for:
- Initial appraisal of time effects on the discounted cashflow
- Subsequent pushback generation
- Subsequent schedule generation
The phase structure created for a given Ultimate Pit is used as the basis for constructing the
OES (a block by block extraction sequence), however the OES is selected to produce the
highest discounted cashflow, whereas each phase has been constructed to represent the
highest undiscounted cashflow for given economic parameters. It will be found that a
combination of blocks in the OES will be close to a phase from the UP. How close they are
will be a function of total size and discounting effects over time.
The Optimal Pit as reported at the UP stage of NPVS may have less blocks than the reported
UP. This is when it determines that the addition of further blocks does not increase the pit
NPV. There is an option, if required, to force the complete the mining of the entire
undiscounted ultimate pit, regardless of time effects.
The structure of phases generated within a UP, this will change the OES. This phase
structure is then used in during subsequent pushback generation. It is important to
understand, therefore, that the phase parameters entered will affect phase generation and the
initially calculated NPV:
- Maximum revenue factor
- Incremental factor
- Type of factor profit, price or mining cost
February 2005
14. Ultimate Pit Results

During the generation of the ultimate pit and phases, the finally determined PHASE and
SEQUENCE numbers are then stored in each block of the ultimate pit block model. Block
model plans and sections depicting these quantities can be produced directly.
Graphical and numeric data reports are generated, collated by both sequence and phases.
Data is collated incrementally (e.g. phase-by-phase) as well as cumulatively. In both cases,
the data is available under the following headings:
Sequence/phase
Economic fields
Rock, ore, waste quantities and stripping ratio
Ore type quantities and grades (if in-situ products requested)
Metal quantities produced (if recovered products requested)
If a sequence breakdown is requested, each record of the data output represents 1% (by
default) of the blocks within the OES. The percentage that each sequence record occupies
can be changed.
If a phase breakdown is requested, each record represents a single phase, along with its
corresponding revenue factor.
The economic fields produced are profit (cashflow), revenue, processing cost, mining cost
and NPV (total DCF according to the OES).
Once an ultimate pit has been generated, bench reserves and grade-tonnage curves may
also be produced of its contents. These facilities are under the Tools menu.
ULTIMATE PIT RESULTS

Pit Optimization Report
Blocks in optimal pit
Total ore tonnage
Total waste tonnage
Profit
NPV estimate
Strip ratio
Ore tonnage by rock type/processing method
Spreadsheet Reports
Reserves, split by bench
Sequence, split by extraction point
Phases, split by pit shell (phase)
Copy and paste into Excel
Visualisation/data export
Model: > Block model with PHASE
Surfaces >Wireframes Ultimate pit or Phases

Slice in DM for bench plans
February 2005
15. Graph Output

Graph output can be produced in 2 ways:
Directly from NPVS
By exporting the results data into Excel, followed by subsequent graphing
As with the examination of the numeric data, NPVS graphs may be broken down by either
sequence or phases. The types of graphs produced directly using NPVS include:
- Line charts
- Histograms
- Stacked histograms
On each of these graphs, the X-axis represents either the sequence or phases, as selected
on entry. The graph quantities on the Y-axis can then be selected as required. This can also
be incremental or cumulative.
When examining the report type tabulated grid data, this may be transferred by Copying and
Pasting into Excel. Once in Excel, it may be graphed in a number of ways. Different
alternative quantities may also be derived in Excel, for additional types of graphs.
Typical graphs, which are particularly useful for analysis of optimisation results, include:
Cashflow/NPV v Phase or Sequence
Ore/Waste tonnage v Price Factor
Strip Ratio v Price Factor
Cost per unit of Metal Produced v Quantity of Metal Produced
Ore tonnage graphs will always have a positive slope. However, the way in which the slope
of this graph changes gives a good indication of the sensitivity of the resource
The strip ratio graph, when considered phase-by-phase, can have both positive and negative
slopes, indicating the following features:
+ve slope => resource supports extra waste stripping
-ve slope => indicates that in-pit material is being converted into ore
Significant deviations in strip ratio often indicate logical surfaces which can be used to
demarcate pushback expansions.
The cost graphs should show 2 lines the average cost/unit metal for each pit, and the
incremental cost/unit metal for each pit.
February 2005
Price Factor v Ore and Value
1.00 25000000
0.90
0.80 20000000
0.70
0.60 15000000
$ Value %
Ore
0.50
0.40 10000000
0.30
0.20 5000000
0.10
0.00 0
0 20 40 60 80 100
Price Factor
Profit $ 95% Max Max Ore
Tonnes v Price Factor
60,000,000 2
1.8
50,000,000
1.6
1.4
40,000,000
1.2
Strip Ratio
Tonnes
30,000,000 1
0.8
20,000,000
0.6
0.4
10,000,000
0.2
0 0
0 20 40 60 80 100
Price Factor
Total Rock Ore Waste 95% Max Max Strip Ratio
Cost Cut-Off v Equiv. Zinc Recovered
1000
800
Cost/t Zn
600
400
200
0
Total Equivalent ZInc
Average Cost Cut-Off Cost 95% Max Max
February 2005
16. Reserve Reporting

Different sorts of reserve reports can be produced, which include:
Summary report within the ultimate pit
Detailed reserve reports by phase
Graph output by ultimate pit or by phase
Grade-tonnage curves
These reserve reports breakdown the contained pit contents according to:
Bench
Economic fields
Rock, ore, waste quantities and stripping ratio
Ore type quantities and grades (if in-situ products requested)
Metal quantities produced (if recovered products requested)
The economic fields produced are profit (cashflow), revenue, processing cost and mining
cost. These quantities can be used to determine other related quantities, such as cost per
tonne of metal produced.
Generally these reports are created by copying the data from the Results grid that is created
in NPVS, and then pasting the data into an Excel worksheet. The phase and sequence
identifiers will often be embedded with the corresponding value reductions (e.g 1 (98%)).
These identifiers and reductions can be separated by using the Text to Columns tool in
Excel.
February 2005
17. Plans And Sections

Plans and sections of the open pit optimisation can be produced in a number of ways:
Directly from the model viewer in NPVS
Contour viewer in NPVS
By exporting the ultimate pit and then using Datamine/Present/GMP
By exporting the ultimate pit/phase surfaces and using the GMP
The NPVS model viewer is a very powerful system for plotting optimisation data in plan,
section or 3D. When the process is entered, the ultimate pit block model will normally be
loaded automatically. Four windows will have been set up automatically a plan, a north-
south section, an east-west section and a 3D view. Each these views will slice through the
ultimate pit block model. Colour/shading legends may be customised to display any of the
ultimate model contents. By default, clipping will also be applied to each window (equal to the
block size in the orthogonal direction). The plane position may also be incremented in steps
equal to this clipping distance. Additional user-defined legends can be easily set up.
Other data may also be loaded into the viewer, including:
String/perimeter/boundary data
Surface data
Point data
The surface data loaded may also have been generated by the open pit optimisation. For
example, if the current UltPit.asc is loaded, then a surface representing the entire ultimate pit
will be loaded. For this sort of data, the following fields may be colour coded:
Elevation
Relative depth
Incremental depth
Bench
To produce a birds-eye view of the ultimate pit, the following steps are required:
1. Load a the UltPit.asc surface.

2. Use Format-3D objects to temporarily disable the block model display.
3. Use View-Direction settings to turn off clipping.
February 2005
18. Case Study Management

The case system allows a tree-like structure of scenarios to be built up, making it much easier
to manage optimisation runs with different settings. The key parts of this case management
system are:
Edit Case to set up case structure and names
Case Notes to record case-specific comments
Select case to select the current case for working
Copy settings each process allows the copying of settings from a parallel case.
The Edit Case facility allows the structuring and naming of individual cases. The branching
of the cases mirrors the structure of the different NPVS processes. e.g.
Economic > Ultimate > Pushbacks > Scheduling

Model Pit
Plans
The hierarchy of cases built up is also represented in the files and directories set up
underneath the current Project Name/Case Studies/..
The Case Notes facility allows detailed descriptions to be recorded along with each case.
The available case structure is always shown diagrammatically in the upper left corner of the
NPVS window. If a particular case is highlighted and then Select Case is used, then this will
be the new current case for on-going work.
In each of the NPVS processes, there is an initial Copy Settings window. This enables the
copying of all settings from a previously executed case. For example, suppose an initial base
case economic model has been set up, and then optimisation work completed. In this initial
base case, no additional dilution has been applied. Suppose now we want to test the effect of
adding, say 5%. Assuming we already have the current economic model settings stored in
the case Economic Model1, with ultimate pit settings in the case Economic Model1/Pit1, the
following steps are required:
1. Use Edit Case to add the following 2 cases:
- Economic Model1 - Pit1

- 5% Dilution - Pit 5%
- 10% Dilution - Pit 10%
2. Click on 5%Dilution/Pit5% and then use Select Case
3. Under Economic Model settings, click on Economic Model1,and then Copy

Settings. All the settings should now be the same as the base case. Now just
change mining dilution to 5%.
4. Generate a new economic model.
5. Advance to the Ultimate Pit process, and under settings, click on Economic
Model1/Pit1, and then Copy Settings. This will copy all slope angles, revenue factor
settings, etc, from the base case.
6. Now generate a new ultimate pit, which reflects the 5% dilution.
The same procedure may be used for any type of alternative setting(s). No optimisation
process can be utilised until a case name has been set up and assigned.
February 2005
19. Ultimate Pit Analysis

Ultimate pit selection will normally be carried out according to defined corporate objectives.
These objectives may be quite different, depending on the mining company involved, the
status of deposit exploration and specific financial constraints for the particular project. Most
objectives involve the assessment of value, of which there are 3 principal types:
1. Cashflow = Revenue - Costs
2. Time Value NPV, Internal rate of return
3. Value in Reduced Risk e,g. short payback, low capital starts
Often the required objective will be the maximisation of value, of the value per tonne mined of
the value per unit of metal produced. NPV-related objectives can suffer from having to have
some sort of schedule, so that an NPV can be devised. At early stages of optimisation,
setting up realistic schedules can be difficult, so it may be better to quickly produce best and
worst case NPVs. If there is a large difference between the NPVs of these two extremes,
then this highlights the opportunity for optimising the NPV through scheduling.
Other criteria that are typically applied include:
Cut-off cost criteria
Maximise metal throughput
Maintain product-oriented blend
The other disadvantage of NPV-related objectives is that the pit limit can become very
sensitive to the discount rate and current price assumptions. Another alternative criteria is a
cut-off cost criteria, such that pit limits are determined in which no ore is mined which will
result in the final product having a production cost greater than the cut-off value. The
specification of this cut-off cost is related to an acceptable risk. Using cut-off cost criteria, at
least for single elements, allows consideration of the pit resource against published cost
curves. Cut-off criteria require the total cost per unit of metal sold to be calculated for a series
of optimal pit shells (phases).
Quite often the finally selected ultimate pit may be the one which is at or near to several
different objectives, for example, an acceptable NPV within a cost/oz restraint, as well as
maximising the conversion of resources into reserves.
Metal throughput can also be very important in terms of loans and other financial
arrangements connected with the project. Product-oriented blends may well depend on the
results available from other pits and/or stockpiles.
The minimisation of risk requires several optimisation runs, with tests of particular pits against
changes in key parameters, such as prices and costs.
In all cases, the analysis of graphical data is important, as is the way in which the deposit
responds to changes in key variables in the decision-making process.
February 2005
20. Existing Pit Designs

It is often useful to re-optimise the pit within an existing pit envelope. This, by definition, will
not change the outer surface of the supplied ultimate pit. However, it will generate both
internal pit phases and an optimum extraction sequence (OES). Reasons for doing this
include:
As a quick evaluation of an existing open pit design
As preparation for pushbacks and scheduling, of an existing open pit design
To use the OES to optimise subsequent schedules
To test phase generation criteria
The way to supply an existing design is simply to identify a surface file (.asc file in NPVS
format). If the supplied .asc file contains multiple surfaces, then the specific pit surface within
the file must also be identified. If the surface to be used has not been generated previously in
NPVS, because its source is a surface from other GMP software, then it must be imported
using the available drivers.
If the objective of pit optimisation is to obtain an ultimate pit which meets corporate goals, as
well as to achieve a practical pit design, then the use of an actual design is one of the final
parts of the overall optimisation/reserve evaluation system, as shown by an example scheme
below:
1. Initial optimisation of block model.

2. Initial design of haul roads, etc, based ultimate pit.
3. Refinement of slope angles from design.
4. Re-optimisation based on more accurate slope angles.
5. Repetition of steps 2-4, until final design completed.
6. Optimisation of phases internal to actual design.
7. Pushback generation and scheduling.
Careful consideration also needs to be given to the slope angle used for optimisation within
the supplied ultimate pit surface. Given that the phases and then pushback expansions may
need to be designed at different working slope angles to the final overall slope angles, it may
make more sense to use the working angles for internal phase and subsequent pushback
development.
February 2005
21. Sensitivity Studies

After or during selection of a base case optimal pit, it is often required to complete a number
of alternative sensitivity runs. There are two fundamentally different types of sensitivity that
can be tested:
A. Different pit optimisations from varying parameters
B. Testing the economic results for a specific pit, for different parameters
The type A sensitivities can be thought of as really testing the deposit, with a view to seeing
what different sorts of pits may be possible. Type B sensitivities are for really testing how
robust a particular ultimate pit, or actual design, can prove under different imposed conditions.
Type A sensitivity runs are often required different costs or prices. In both of these cases,
there are 2 alternative ways of getting sensitivity results:
1. Define separate economic model cases, with different cost/price values.
2. Use the phase-revenue factor tools when generating the ultimate pit, so that
each phase automatically represents a parameter. These phases can therefore
represent different:
o Profit factors
o Price factors
o Mining cost factors
Other types of sensitivity analysis can be run using the case management system. For type B
sensitivities, normally a pre-existing ultimate pit surface will be used. For comparison of
results, a spider diagram is often used. This is usually shown with %variation in the varied
parameter on the X-axis, and variation in overall pit measure on the Y-axis. This Y-axis
value could be total cashflow, NPV or internal rate of return.
The IRR can generally only applicable for new projects, where there is a capital input as the
balancing negative cashflow. If, however, the change IRR is plotted on the Y-axis, the slope
of each line can give measure of sensitivity.
Base Case
30% Price
25%
Variance in IRR
20%
15%
10%
5% Capital
Operating
0% Costs
-5%
-10%
70% 90% 110% 130%
Variance Parameter
February 2005
22. Pit Surface Export

After an ultimate pit has been generated, there are 2 separate pit surface files generated:
Ultpit.asc - Just the (maximum cashflow) ultimate pit
Phases.asc - Surfaces for each pit phase (shell)
These may be exported in two ways, which will lead to different types of wireframe files
being created:
1. The Surface Viewer can be used to view either of these surface groups in a dynamic 3D
window. When in the viewer, the loaded surfaces can be exported. If the UltPit.asc is
loaded, then only one surface will be exported. If exported to Datamine, the surface will
be converted into a surface wireframe model a corresponding pair of triangles and
points. The ultimate pit surface represents the ultimate pit itself, as well as the original
topographical surface outside of the ultimate pit itself. This type of surface is completely
3D mesh, effectively smoothing the edges of the ultimate pit block model.
If the Phases.asc file is loaded, then if a surface export is required, the particular surface
(phase) for export must also be selected. Alternatively, the Edit Surface File facility can
be used (from the Tools menu) to make a new separate .asc surface file, which is either a
part of or a combination of a number of existing surface files. Once the new .asc file is
created, it may then be exported into other GMP formats.
2. The Model Viewer can also be used to export a surface. If a UltPit.asc is loaded into the
Model Viewer, a surface is created which is comprised of flat surfaces only, effectively on
each berm surface. This allows very clear birds eye plans to be produced, colour coded
in the same way as a block model. When such a surface is loaded, 3 extra attributes are
also created: Bench, Relative Depth and Incremental Depth. If the Export Surface
command is now used, any or all of these attributes may also be exported with the
triangle data.
Once exported into the GMP, there are a number of uses for such surfaces:
Horizontal slicing to form horizontal bench contours
Vertical slicing for pit cross-sections
Birds Eye views of ultimate pits/phases
Evaluation of block model contents
To help with processing of the pit surface, a set of pit boundary points from the NPV
Scheduler can also be output. These can be used to separate the exported surfaces into
separate components inside and outside of the pit limits.
The exported ultimate may also be used in the GMP software (e.g DATAMINE) to create a
wireframe envelope of the pit envelope.
February 2005
23. Optimisation and Practical Open Pits
Theoretical pit => practical pit
Additional ramp tonnage = Density x Width x Height

2 x Gradient
Critical factors :-
- Location (up to 10% of pit value can be lost)
- Width
- Gradient
- Smoothness
Once an initial ultimate pit has been produced by optimisation work, a proper designed pit is
required, which incorporates haul roads, berms and different bench/berm configurations.
A rough estimate of the additional tonnage required for the addition of haul road into a pit
(without a haul road), and its associated mining cost, will quickly demonstrate the care
required with its placement.
Smoothing of the initial optimal pit shape, followed by pit evaluation of a series of quick and
approximate road designs, will allow a much better understanding of the sensitivity of different
areas of the haul road. Using the best result, a more detailed ramp design can be made,
along with toes, crests and berms. Measurement of the resultant overall slopes in different
directions will probably yield different values to those used in the original optimisation.
Therefore, time permitting, the best approach is now another run of the optimisation, using the
revised (more accurate) slope angles. This can then be used as the basis for further more
detailed design. Several iterations may be required, but may produce a significant
improvement in economic value of the final design (4).
This effect of refining the design, and then re-optimising with improved (more realistic) slope
angles, is demonstrated in the diagram below.
Target
X
$ Value
Optimisation
Pit design
LESS Practicality Index MORE
February 2005
24. Pushback Generation
Creation of expansions within defined ultimate pit
Initial control by ore quantity and depth
Access space and minimum size
The ultimate pit within which pushbacks are created can be defined in one of two ways:
1. By taking the maximum ultimate pit produced previously in the same preceding case.
2. By defining a specific block number within the previous ultimate pits optimum
extraction sequence.
The next main control is to define the maximum number of pushbacks that can be generated.
In addition to this, an option can be enabled, which ensures that the pushbacks will definitely
reach the final (supplied) ultimate pit shell.
For each pushback, the minimum quantity of ore is specified, as well as the maximum depth
of each pushback. The same parameters can be defined for all pushbacks, or different
parameters per pushback.
Another option defines whether each pushback must forma contiguous volume i.e. can the
same pushback be made of more than one physical volume. Access space (width) can also
be defined, to ensure that each pushback meets minimum width requirements. In addition,
the minimum size of any individual pushback can be defined, in terms of the number of parent
blocks.
Once these main parameters have been defined, the pushbacks can be generated. The data
resulting from this can be summarised as:
A new ultimate pit block model, which now has a PUSHBACK value in each block,
adding to the PHASE field already there from the ultimate pit generation.
A set of surfaces for the base of each pushback (Pushback.asc)
Results detailing the contents and cashflows of each pushback
As with the ultimate pit generation, both the pushback block model and associated surfaces
can be exported. The different pushbacks can be viewed in plan or section.
There is also a boundary correction factor which affects the smoothness of the resultant
pushbacks.
For large pits, pushbacks might logically be designed so that they fit in with business risk
periods. The resultant pushback width is very important, as it will affect how much waste has
to be moved to give a minimum ore exposure. In addition, wider pushbacks can result in
greater blasting efficiencies, and more room for fleet movements.
During scheduling, average grades per bench are utilised, so the pushback configuration will
also affect the database of grades available for specific grade-related targets.
February 2005
PUSHBACK CONTROL
Ultimate Pit Definition
If selected from particular phase or existing design:
Select that surface as ultimate shell
Re-optimise within that, at required price levels
OR
Limit pit by block number:
Determine from graph/results of sequence analysis
Main Options
Maximum number of pushbacks to generate
Last pushback to reach final pit shell
(if not, pit may be smaller by NPV considerations)
Access space
Minimum blocks in a remnant
Pushbacks from contiguous volumes
Limit pit to block
Control
Define variables
For each pushback
Variable quantity
Maximum depth
Advanced
Boundary correction factor
1,2,3,4
Smoothing
Model Data Generated

PHASE
SEQUENCE
PUSHBACK
Adjust
Multiple tabs for each pushback
Legends available:
Incremental depth (from last pushback)
Relative depth (from initial topography)
Bench
Elevation
Objects > Adjust Pb. K
Polygonal boundary > Apply:
Insert boundary > apply
Edit | Copy Trace
Region Type:
Global limits Confines entire pushback and limits depth
Local limits Limits depth of pushback
Local include Modifies to include blocks
Local exclude Modifies to exclude
February 2005
25. Pushback Customisation

There are 4 different ways in which pushbacks may be customised (and/or generated):
1. Importing of existing pushback surfaces, from external sources, such as those which
have been designed. The Surface Editor can then be used to combine a number of
surfaces into a single .asc multiple surface file.
2. Use of Surface Editor to pull out a set of selected phase surfaces from a generated
Phase.asc file.
3. Use of Adjust tool to edit previously generated pushbacks.
4. Use of Pit Partitions to split up an existing set of pushbacks. This could even be
used to generate a complete set of pushbacks, starting from the ultimate pit.
The Adjust tool offers the following facilities:
View created pushbacks individual windows
Customise by controlling polygons
Different kinds of polygon limits
Once the pushbacks have been generated from the parameterised ore controls, each
pushback may be viewed in an Adjust program, containing separate plan windows. The
pushbacks can then be customised by the definition and/or import of controlling polygons.
These polygons can be used to control pushbacks in one of the following 4 ways:
1. Global limits - Confines entire pushback and limits depth
2. Local limits - Limits depth of pushback
3. Local include - Modifies to include blocks
4. Local exclude - Modifies to exclude
Separate groups of limits can be defined for each pushback, and then enabled/disabled
during subsequent pushback generation. These facilities allow considerable control over how
pushbacks are generated, and the pit subsequently evolved. The procedure is iterative, such
that the pushback limits and parameters can be refined until the required pushback
configuration is achieved.
These limits can also be used to control the ultimate pit shape. In particular, local limits can
be used to remove excessively small drop-cuts, as well as to stop small drop-cuts on
individual pushbacks.
The final phase generation inside the finally decided ultimate pit (existing surface) may well
be done with a different set of angles to those used in the optimisation. As the phase angles
will control the generation of pushbacks, it may make more sense to use typical working slope
angles for this final phase generation.
Pushbacks can also be created, or modified, by the application of pit partitions, These are
pit surfaces which have been split up into separate parts (partitions), by definition of a bench
'footprint' and accompanying slopes, within a larger ultimate pit envelope or already-defined
pushback surface.
February 2005
26. Project Overview

For most optimisation projects, typical presentation data that can be prepared for reporting
purposes will often include:
Model Input and Economic Modelling
Economic parameters
Model import summary
Model validation data (comparison with resource evaluation)
Economic model settings
Plans, sections and contours of economic model
Example check value calculation
Ultimate Pit Generation
Ultimate pit settings
Tables of main results
Graphs of optimisation results:

o Cashflow/NPV v Phase or Sequence
o Ore/Waste tonnage v Price Factor
o Strip Ratio v Price Factor
o Cost per unit of Metal Produced v Quantity of Metal Produced
Summary of sensitivity analysis results, spider diagrams
Summary of pit designs refinement of slope angles
Discussion of pit analysis => Ultimate Pit selection
Plans, sections and Ultimate Pit
Bench reserves and grade-tonnage curves of Ultimate Pit
Pushbacks
Pushback settings + controls applied
Plans/sections of generated pushbacks
Tables of pushback contents
February 2005
27. Terminology
Maximum number of benches between benches mined being at the same time between
Bench Lag
consecutive pushbacks
A particular branch defining a specifc set of programm settings and results within the
Case Study
NPVS case management system.
When local pit limits are defined, they may have costs associated with them, so that
Conditional Pit Restrictions removal of these restrictions is conditional upon the optimal pits generating suffiicient
profits to justify their removal.
Software drivers used to import and export data to and from mining software and CAD
Data Source Drivers (DSD)
systems.
This is the reduction of a cashflow, normally pertaining to a specific year, based on the
Discounting elapsed time from the start of the project. For a given discount rate, d and a cashflow V,
and number of years N. The discounted cashflow (DCF)=V/(1+d)^T
Cut-off calculated by NPVS, which is determined from the equation revenue/t =
Economic Cut-off grade processing cost per /t. For a constant recovery, Cut-Off = (1+Dilution)*Processing
Cost/[(Price-Selling Cost-Additional Processing Cost/metalunit)*Recovery]
Block model in which each block has had an economic value calculated.
Economic Model
VALUE=REVENUE-COSTS
Global limits Polygon restricting the extent of pushback generation
Part of NPVS which allows truck haulage hours to be calculated and then utilised as an
Haulage Analysis
additional scheduling target.
A lookahead value for a block B is a discounted sum of profits to be had from B and a
set of blocks contained in a downward looking cone with a vertex at B. These
Lookahead Value
lookahead values, together with the phases, are used in NPVS to determine the optimal
extraction sequence.
The OES is the sequence of blocks within the ultimate pit, that will yield the highest
Optimal Extraction Sequence (OES) possible discounted cashflow. It is affected by the discount rate, the ore production rate
and the internal phases which have been generated.
After generation of the ultimate pit, alternative internal phases (or pit shells) are
generated, each of which may be considered an optimal pit corresponding to 'worse'
economic factors than those of the supplied 'base case' parameters. The economic
Phase
factors for these phases may be increments of the each blocks' profit value, the metal
price(s) or the mining costs. Phases may also be generated 'beyond' the ultimate pit
my use of a maximum revenue factor greater than 100%.
These are pit surfaces which have been split up into separate parts (partitions), by
Pit Partitions definition of a bench 'footprint' and accompanying slopes, within a larger ultimate pit
envelope.
Once an ultimate pit has been defined, or particular sub-shell of the ultimate pit,
Pushback separate physical volumes (pushbacks) can be generated, which normally represent the
limits of logical mining expansions.
In NPVS 'sections' are physically separate volumes, which are separated by sub-
Section horizontal surfaces, used for the purpose of splitting up a block model framework into
separate slope regions.
A slope region is a physical volume within the overall block model framework, to which a
Slope Region particular group of bearings and corresponding corresponding slope angles apply.
These slope regions are created by defining limits horizontally by supplied perimeters (X-
Y partitions) and/or vertical divisions created by supplied separating surfaces.
In NPVS a 'surface' is three-dimensional surface model. In NPVS format it usually has

a .asc file extension. A single .asc file can contain a number of physically different
surfaces, each of which will have an identifyign index number. Surfaces are analgous to
Surface wireframe surfaces or DTMs in most mining software systems, can be created from
these models by use of the import drivers. Surfaces may also be similarly exported as
DTMs. In NPVS, surfaces can data types which include topographies, slope division
surfaces, mining surfaces, pushbacks and pit designs. There is also a surface editor to
allow the combination or separation of surfaces.
This is a method of reducing block values according to vertical depth, by a discounting
Top-down discounting method, so as to yield an optimal pit with a better overall DCF than pits which
(individually) are based on maximisation of pure cashflow.
This is usually the maximum cashflow optimal pit (based on the supplied economic
Ultimate pit (UP) model) generated during the 'Ultimate Pit' optimisation. Internal phases may also be
generated inside the ultimate pit. Further shells may also be generated beyond the
ultimate pit, if a revenue factor greater than 100% is used for phase generation.
Block values are calculated as basically being VALUE=(REVENUE-PROCESSING
Value COST-MINING COST). For waste blocks the value will therefore simply be = -MINING
COST.
These are separate physical areas, when viewed in plan, which are used in the breaking
XY Partitions up a block model frameworked into separate slope regions. They are created by
defining or importing perimeters (polygons).
February 2005
February 2005

Strategic Open Pit Planning and Optimisation

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Strategic Open Pit Planning and Optimisation

Hochgeladen von

Copyright:

Verfügbare Formate

STRATEGIC OPEN PIT

PLANNING AND OPTIMISATION

Presented by Richard Gawthorpe

An overview of the NPV Scheduler system is shown in the flowchart overleaf.

The main components of the NPV Scheduler can be summarised as:

NPV Scheduler - Overview

Import Geological Block Model

Metal prices/ Mining and

Slope Settings Pit Shell Optimisation Pit Limits

Pit Result Model:

Select Particular Ultimate

Cashflows/ Scheduled Block

Lerchs and Grossmann publish (1964) mathematical method for

Three-D (1985) implementation on a range of computers. Single pit

Other optimisation methods

Four-X, NPV Scheduler

3. Economic Parameters Preparation

Develop spreadsheet model

Include all costs that stop when mining stops

Sometimes useful to think in terms of contractors

Time and replacement capital costs

Extra ore mining costs must be added to processing costs

CATEGORY PARAMETER UNIT VALUE

Economic Parameters' Preparation

Mining Costs' Preparation

Processing Costs and Parameters

Other Operating Parameters

Block Model Preparation

4. Block Model Preparation

Tonnage direct or by calculation with DENSITY field.

Costs initially calculated for REFERENCE block

Mining cost adjustments can be supplied as:

Processing cost adjustments can be supplied as:

Rock type mining CAF = Mining cost for rock type

Potential Data Types

Input Model Report

5. Block Model Import to NPVS

Grades or metal quantities can be supplied.

Tonnages can be supplied directly or calculated.

Rock type field (numeric)

Mining and/or processing cost adjustment factors (optional)

Other controlling surface or attribute fields (optional)

After model import, the report of model contents should be inspected.

Block Model Rock Types

6. Economic Model Generation

Cost calculations required?

Data for reporting

Mining costs, factors and adjustments

For each processing method, the following parameters are required:

- Total mining cost

7. Cut-Off Grades and Block Allocation

Revenue > rehabilitation cost

Sum of (grades / cut-offs) > 1

All grades > cut-offs

Any grade > cut-off

The calculated cut-off grade is determined by the formula below:

Cut-Off = Processing Cost x (1+Dilution)

Reasons for defining cut-offs may include:

Pre-prepared Mining Factors Advantages

- Can be engineered in great detail,

Pre-prepared Mining Factors Disadvantages

NPVS Mining Factors Advantages

- Easy to apply and vary for different scenarios.

1. Potential Revenue (PR) = (Price-Selling Cost)RecoveryGrade*Mining Recovery