SMG Minex Software Overview

Minex Software
Minex is a comprehensive mining package suited to planning and control of coal and
other strata bound deposits; including coal bed methane, bauxite, beach sands and
phosphate. Minex covers the range from Life of Mine (LOM) planning to daily
production. Minex is used by mining companies such as:

• PT Kaltim Prima Coal, East Kalimantan (KPC)

• Cerrejon Coal, Columbia
• Rio Tinto Coal Australia (Hunter Valley)
• BHP Billiton (Ingwe Collieries) South Africa
• Sasol Coal, South Africa
• Anglo Coal Australia
• Jordan Phosphate Mines Company (JPMC), Jordan
• Central Mine Planning Design Institute (CMPDIL), India and
• Tata Iron and Steel Company (TISCO), India.

Minex covers the full planning process of exploration, modeling, ranking, design and
scheduling, both short and long term.

The main components of Minex are:

1. Horizon - Geology

Horizon is our geological evaluation system that includes an integrated borehole database
and 2D gridded modeling package. The database is user defined and allows storage and
manipulation of survey, laboratory and geophysical data. A key feature of the database is
that the coal seam interpretation is independent of the data and thus the seams can cross
sample intervals. Compositing can be completed when the data is displayed or modeled.
This independent interpretation allows the geologist to have multiple interpretations of a
deposit (e.g. bulk mining A or selective mining A1A2 as seen in the figure below).

Figure 1:
ASH
Prior to seam modeling, Minex extends
ROCK
short drill holes to contain all seams using a
86 SS
combination of techniques such as missing
seam interpolation, seam splitting and
10 CB A1
85 SS A father son. These techniques allow shallow
holes to be deepened to the full stratigraphy
12 CB A2 and thus maximize the value of the
70 MS collected data.

1
SMG Minex Software Overview

Horizon uses trend surfaces, inverse distance or kriging to build models. Models can be
built automatically for up to 200 seams and virtually an unlimited number of quality
variables (ie: ash, moisture, energy, sulphur, fluidity etc). Horizon incorporates strong
model arithmetic to allow creation of derived models such as strip ratio, diluted or run of
mine models (ROM) and operating cost per tonne (or margin per tonne) models. SQL is
available on the borehole data and the model. SQL allows creation of models based on
functions such as

If thick < 3 dilution = 0.1 else dilution = 0.12 etc.

Resource reporting can be based on polygon or, depth and can incorporate JORC
classification based on a competent persons distance from data point criteria.

The standard Horizon package allows modeling of vertical faults. The advanced Fault
modeling option models inclined faults: normal, reverse or thrust. The Minex fault model
technique moves the borehole data into un-faulted space, models the seams then moves
the model back to faulted space. The technique offers advantages in modeling barren
zones and understanding the faulting.

Figure 2: Horizon plan and section plot and multi seam faulted model

2
SMG Minex Software Overview

1.1 Coal Quality Database

The Minex coal quality database (CQD) stores data with differing float, sink density and
sizing. Data can be accumulated into reports and/or plots showing target ash or plant
specific gravity as an example. Data can also include froth float information. The coal
quality data is linked to the Horizon database so information can be modeled and used in
resource reports or mining studies.

Figure 3: Coal Quality database plot

3
SMG Minex Software Overview

2. Apollo – Design and Scheduling

MINEX Apollo overlays an open pit design of benches, strips and blocks on a Horizon
geological model and provides the engineer with schedulable volumes and quality.
Apollo includes options to convert a resource to a reserve using suitable combinations of
the following:

• Seam dilution on thickness or percentages

• Seam combining (including non separable partings) and
• Seam wasting on thickness or quality criteria
• Adjustment from insitu to product or ROM moisture basis.

Where the process of creating reserves is complex this process can be automated. This
avoids errors, ensures reproducibility and speeds processing. This automation of the
insitu to ROM process can be password protected so a “super user” can maintain control
on the factors, while other users realise the speed and reliability benefits.

Reserve blocks can be edited using a block by block approach or using a SQL. The SQL
approach allows major editing based on attributes. A simple SQL such as

TONNES = TONNES x 0.98

could allow for coal rib loss in a dragline operation.

Apollo scheduling is fully integrated with the reserve database. Scheduling options
include:

1. Detailed scheduling using multiple equipment items and multiple pits or deposits
2. Simple single item equipment scheduling where other activities are ignored. This
is most suited to a large waste unit such as a dragline. In this case other activities
such as coaling and drill and blast is assumed to occur.
3. Target scheduling is suited to targeting ie. 1 million tonnes per annum from the
deposit. All other activity is assumed but can be reported from the schedule.
Target scheduling is an excellent aid to deposit understanding.
4. The Apollo database can be used as the input to Maximiser LOM scheduling (see
section – Maximiser – LOM Scheduling).

All scheduling can be displayed graphically during or post scheduling. This aids
schedule validation and site “selling” of the schedule. As the Apollo system is integrated
with the geology model, any schedule time (ie. 6 months, 12 months 5 years) can be
converted to a mined out surface and used for dumping studies. Apollo reserve and
schedule reporting output files are in CSV format which can be readily loaded into a
spreadsheet.

4
SMG Minex Software Overview

Figure 4: Apollo Schedule display showing areas drilled, charged, blasted and
excavated

Drilled

Blasted
Drilled

Charged

Hatching indicates drill & blast status Mid period block is now part charged

2.1 Spoil Design

Minex includes three tools for the planning of dumps. These tools are:

1. A dump designer (maximum spoil) which uses a terrain surface and a footprint
polygon to build a dump of a defined size (in cubic metres) or at a defined surface
elevation. This module is ideally integrated with the surface created from the
Apollo scheduler.

2. Cross sectional editing of a dump surface is suited to dozing dumps into a final
landform. This module is ideal when the push direction is obvious such as a
highwall or a dump edge. In these cases a series of sections parallels the push
direction and mass is balanced with allowance for swell.

3. For 3D problems when the push direction is unknown LP spoil regrade is ideal.
This module uses a LP to find the best solution to leveling dumps. The ideal
outcome is measured as the minimum of the product (distance pushed x volume
pushed). Spoil regrade is ideal for dragline spoil heaps. Push can be controlled
by final slope, final elevation and optionally drainage strings (to allow water
flows). Input is an initial surface and the output is a final surface and push
direction information. This data allows creation of plans and push arrow plots
which at some sites are directly loaded into the dozer cabin computer.

5
SMG Minex Software Overview

Figure 5: Maximum Spoil design (blue and yellow) over mined surface green.

Figure 6: Sectional editing for rehabilitation etc.

6
SMG Minex Software Overview

Figure 7: Spoil regrade. Before and after & push arrows, arrow length indicates volume.

2.2 Drill & Blast

The drill and blast package sets out blast patterns. The pattern can be square or staggered
and can incorporate pre-split rows, vertical or inclined holes. Holes can be dumped into a
survey system or on board drill systems (ie. Aquilla or Leica). The pattern, row or
individual holes can be edited using a blast pattern explorer system. The drill and blast
system is integrated with the geology model and survey packages so holes can be drilled
to or stand off a seam. Blast zones displayed as cylinders around the hole allow the user
to locate problem holes (too close or too far from a free face) and relocate these holes.

7
SMG Minex Software Overview

Figure 8: Typical blast pattern layout with blast explorer system

2.3 Survey System

The Minex survey system uploads data directly from a survey instrument and uses a
wizard to walk the users through the load and reduction process. The end point of this
process is a set of XYZ labeled survey points which can then be edited using the standard
Minex core string functions. Typically the surveyor will build period surfaces and use
differences between periods to generate period volumes. Strings can be designed in
Minex and uploaded to the survey instrument for field setout.

2.4 Truck

The truck simulator estimates truck travel time based on a loading unit and a haul road.
The roads are based on a set of XYZ strings. These strings provide length and grade and
the user assigns attributes such as rolling resistance to the string. Strings are built into
haul roads and the return trip can be an automatic reversal of the strings or an alternate
route if required. The truck is then “traveled” along the road using the truck rim pull,
retard and braking data. Outputs include travel time and production achieved with
integer numbers of trucks (1,2,3,4 etc) for that haul.

8
SMG Minex Software Overview

3. Eclipse – Underground Design and Scheduling

Eclipse is an underground room and pillar and longwall design and scheduling package.
Eclipse designs are based on centerlines. Centerlines are assigned attributes such as pillar
spacing, width and cross cut width to ease the design process. Pillars are generated
automatically from these attributes. Reserves are generated using the Horizon geology
model with volumes, tonnes and coal quality all read from the model. Data can be
manipulated to include costs or production factors.

Eclipse uses the network of centerlines as a schedule network with activities and
precedences determined from the centerline attributes and their connections. Scheduling
uses a mixture of a critical path network method (CPM) and manual control (if
necessary). Scheduling can be as automatic or as manual as the user requires.

Figure 9: Eclipse Interactive Schedule display & period plot.

FI
1

FI
2

1
FI

TE
GA
3

1
LW

F
2
I4
TE
GA
D RIF T
2
LW

FI
3
TE

5
GA
M
A IN
/1 3
LW
4
A TE
WEST1/R G

FI
6
4
LW
MAINS/1

5
TE
GA
WEST1/2
M 5
A IN
LW
/2

6
A TE
G
M
MAINS/2

AI 6
WEST2 N /3 LW
EA
ST
1/
L2 E7
G AT
M
A IN
/4 7
LW
EA
ST
MAINS/3

WEST3 1/
2
8
M TE
A IN GA
/5
8
LW
EA
WEST4 S T1
MAINS/4

/L MA
1 IN 9
/6 TE
GA

EA
ST 9
1/ LW
3
M
WEST5 A IN
/7

E AS A
G
T1
/R
MA
IN
/8
WEST6/2
MAINS/6

M
AI
N /9
WEST6/L

9
SMG Minex Software Overview

4. Pit Optimisation

MINEX’s pit optimiser, adapts the Lerchs-Grossman (LG) optimisation techniques to

coal or layered type deposits. PitOPT is integrated with the MINEX horizon model. For
optimisation a 3D block model is built from the 2D model with minimal user input.
Typical inputs are mining costs, washing costs, pit slopes and revenue. The output of the
optimizer is a 2D surface or grid which can be used as a reserve shell for pit design.
Running the optimiser at different cash profit margins provides a set of maximum profit
pits (for those margins). These “nested” pits are a useful guide for design and scheduling.

Figure 10: Nested pits should be mined in sequence to maximize NPV.

5. Maximiser

Maximiser is a LOM scheduler system that handles issues with mining, blending and
trucking. Maximiser considers strategic or LOM decisions. For example blending A and
B today may make tomorrows (or next years) blend difficult. By discounting (say 10%)
the profit from future periods and maximizing NPV the correct option can be selected.
Decisions on trucking and dumping are similar. Short dumping has a low cost but results
in longer future hauls and requires extra capital (more trucks). The best choice between
short haul today and future longer hauls is made based on net present value (NPV)
maximisation.

Maximiser is used by mining companies to evaluate strategic issues that determine

business value. Maximiser can be used in coal or metals and has been used to evaluate
issues such as:

1. Market mix over time. Given a mine what specification can be shipped say for
the next 20 years?
2. Given a set of N mines what is the future market mix?

10
SMG Minex Software Overview

3. Given a mine or set of mines where is the best location for a washery or mill?
4. Given a deposit mined with a dragline and truck shovel what is the best working
horizon or balance between truck shovel and dragline. This issue is important in
coal mines as the dragline is the lowest cost overburden removal equipment.
However as extra overburden is allocated to the dragline, its costs increase due to
extra rehandle and the coal uncovered rate also falls.
5. Given a mine what is the capital expansion or replacement schedule.

Maximiser can readily handle the large strategy issues such as blending from multiple
mines or plants. Maximiser is typically an open cut tool but a set of underground blocks
can be easily used in Maximiser as an alternate blending source.

Figure 11: Maximiser output is graphed in Excel to compare NPV of options.

COMPARISON CUM NPV

$2,500

$2,000
Cum NPV ($mill)

$1,500

$1,000
Option 1
$500 Option 2
Option 3
$0

-$500
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Project Year

11