A Non-Linear Programming Model for Open Pit Mine

Equipment Selection
A Aghajani Bazzazi1, M Osanloo2 and B Karimi3

ABSTRACT
In mining industries, materials loading and handling costs represent a significant component of the
total (capital and operation) costs. A variety of methods have been applied in surface mine equipment
selection in order to select the suitable loading and haulage system with the lowest total cost. The
optimal size and number of equipment calculation are very important task for mining engineering
because loading and haulage equipment are major proportion of open pit mining cost. Capital and
operating cost are considered in objective function with respect to interest rate, insurance and tax
in non-linear form. Shovel and truck annual productivity constraint, fleet matching constraint and
number of truck per shovel constraint are incorporated in this optimisation problem with continuous
variables. This formulation will yield total cost minimisation per hour and good accuracy compared
to previous methods.

INTRODUCTION
Equipment selection is one of the most important aspects of open pit design. Mining costs are
mainly affected by the number and capacity of equipment. Equipment selection for open-pit mines
is definitely a major decision which will impact greatly the economic viability of an operation. The
purpose of equipment selection is to select optimum size and number of equipment with minimum
cost (Osanloo, 2007).
Loading and haulage cost are major proportion of open pit mining costs. Truck haulage costs are a
substantial proportion – typically some 38 per cent compared with 15 per cent for drill and blast and
12 per cent for truck loading, see Figure 1 (Gregory, 2003).
A review of relevant literature indicated that operational research optimisation techniques currently
in use have serious limitations. These techniques lacked flexibility and often were invalidated by their
inability to cope with a large number of variables, constraints and uncertainty, which are a natural
part of the process of removing overburden and ore extraction (Jayawardane and Harris, 1990).

Typicalopenpitminingcostbreakdown
Drilling
5%

Management& other
supervision 12% Blasting Drilling
6% 10%
Blasting
Loading Loading
Ancillary 12%
Haulaging
equipment
17% Ancillaryequipment
Management&supervision
other

Haulaging
38%

FIG 1 - Typical open pit mining costs by activity (Gregory, 2003).

1. PhD Student, Department of Mining and Metallurgical Engineering, Amirkabir University of Technology, Tehran, Iran. Email: [email protected]
2. Professor, Department of Mining and Metallurgy Engineering, Amirkabir University of technology, Tehran, Iran. Email:[email protected]
3. Associate Professor, Department of industrial Engineering, Amirkabir University of Technology, Tehran, Iran. Email: [email protected]

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 303
A AGHAJANI BAZZAZI, M OSANLOO AND B KARIMI

The solution for the equipment selection problem has its own history. A fundamental concept is
presented in the investigations of Singhal (1986) and Singhal and Fytas (1986). These investigations
place a strong emphasis on the formulation of technological and geomechanical criteria of the
production equipment selection that have to meet the specific requirements of the mining method
used and geomechanical status of rock masses. A thorough analysis of the operating costs of the
equipment utilisation is done as a finalisation of the selection procedure.
The next stage of the problem solution is characterised by investigations dedicated to the role of
the economic factors in equipment selection. A result of these investigations is the development of
the life cycle cost (LCC) analysis (Rao, Reddy and Sathaiah, 1989). The analysis provides a well-
grounded approach to the equipment selection problem. It is based on the total costs distribution
over useful life that overcomes the disadvantage of the analysis of the purchase costs as the only
economic criterion of the equipment selection. Investigations in this direction were also pursued
by Hrebar (1990) who developed an approach to dragline selection procedure in strip mining.
Original elements of his approach are undoubtedly the implementation of ‘production rate – bench
geometry’ relationship, which plays an important role in the methodology of equipment selection.
The selection procedure uses discounted cash flows analysis to account for the time distribution of
the capital and operating costs of the equipment. A shortcoming of the approach lies in the absence
of a mathematical optimisation technique, which could provide a scientifically grounded optimal
solution for the equipment selection problem.
The economic factors are also well investigated by Bozorgebrahimi et al (2003) with regard to the
selection of trucks. Investigations based on the implementation of optimisation techniques are made
by few researchers. Usually linear programming and mixed-integer programming techniques for
developing optimisation models of equipment selection were used (Maitra et al, 1994 Reznichenko
and Valuev, 1994 Cebecoy, Gozen and ahsi, 1995).
Various mathematical techniques are also used in seeking solution such as the genetic algorithm
(Haidar and Naoum, 1995 inchun and oudi, 2004), knowledge-expert systems (Bandopadhyay
and Venkatasubramanian, 1987 Ganguli and Bandopadhyay, 2002, Denby and Schofield 1990) and
decision-making techniques (Denby, Clarke and Schofield, 1990 Samanta, Sarkar and Mukherjee,
2002 Bascetin, 2004 Bandopadhay, 1987 Aghajani, Osanloo and Soltanmohammadi, 2008
Aghajani, Osanloo and Karimi, 2009).
The paper offers a mathematical model of the loading and haulage equipment selection in open pit
mines that is developed as a non-linear programming problem. The model provides the equipment
selection optimisation as part of the production scheduling problem based on minimising cost. The
main objective of the study is to assist open pit mining company in the decision-making problem
for selecting the right size and number of equipment and compare their decision with the proposed
model outcome.

THE OPTIMISATION MODEL

This section involves the use of optimisation techniques to define the size and number of the
equipment fleet that would produce the minimum total cost of the operation. To process this sort
of problem, it needs to be presented synthetically, which involves deriving a set of input parameters
required to generate a set of desired outcomes.

OBJECTIVE FUNCTION
Objective function is a function of the numeric variable that needs to be optimised. In this case,
minimising the total cost of owning and operating the various types of equipment that operating
in the mine is objective function. It is a function of the number of equipment to be utilised and
their cost. In this optimisation problem, we want to optimise four independent variables (number
of shovel (x1), shovel capacity(x2), number of truck(x3) and truck capacity(x4)). For this purpose, at
first the relation between shovel capacity and costs of owning shovel should be recognised. Based on
Table 1 and Figure 2, this relation between capital cost and shovel capacity is as
Ssh ( ) 1686.8(x2)2 + 192 273(x2) (1)
where
Ssh is the purchase price of hydraulic shovel
x2 denotes to shovel capacity

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 304
 A NON-LINEAR PROGRAMMING MODEL FOR OPEN PIT MINE EQUIPMENT SELECTION

TABLE 1
Data of hydraulic shovel costs (CostMine, 2008).

Bucket capacity of hydraulic shovels Capital cost ($) Hourly operating costs ($/hour)
cubic meters) (US dollars, 2008) (US dollars, 2008)
2.3 604 000 70.05
3.4 679 000 89.66
4 824 000 95.95
5.2 969 000 92.47
5.3 739 000 115.25
5.7 994 000 118.82
6.3 819 000 129.63
7 1 262 000 136.41
8 1 648 000 206.92
8.4 1 55 3000 202.55
10 1 848 000 232.18
11 2 388 000 276.99
14 3 676 000 380.87
14.5 4 346 000 379.83
15 3 097 000 408.59
16.1 4 396 000 444.63
18.3 4 096 000 448.73
23 4 955 000 531.03
26 6 144 000 636.98
30.6 7 592 000 781.82
33.6 6 893 000 798.6
39.8 9 291 000 1035.12
42.8 12 887 000 1124.03

FIG 2 - Relation between capital cost and shovel bucket capacity.

It is notable that as mentioned before, we want to calculate unit ownership cost ( /hour) instead of
capital cost ( ) in objective function. The unit ownership cost is considered as a fixed charge per hour
which the owner has to pay irrespective of operating conditions (Hartman and Mutmanski, 2002).
The unit ownership cost of excavation equipment has two components
1. depreciation component, and
2. interest component.

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 305
A AGHAJANI BAZZAZI, M OSANLOO AND B KARIMI

The simplest method for calculating the depreciation component (Dsh) is the straight-line formula
having the expression

D sh S sh - R sh (2)
#H
where
Rsh is salvage value of excavation equipment
N is useful life of excavation equipment in years
H is hours of service of excavation equipment per year
The interest component can be assessed by the average investment method. The average investment
method has two components one representing the wearing out of the equipment and another
representing the return of the money the company has tied up in the equipment while it is operating.
The method assumes straight line depreciation and a percentage of the annual average investment
to cover interest, taxes and insurance (Runge, 1998). The basic form of interest component (Ish) is
expressed as follows
S sh ^r + i + ph^^ + 1h /2h (3)
I sh
#H
where
i is the interest rate on capital of excavation
r is the insurance rate of excavation equipment
p is excavation equipment taxes
So the unit ownership cost per hour consists of the depreciation component and the interest
component, therefore it is expressed as

Co sh
Dsh + Ish (4)

Homogenous fleet of loading-haulage equipment was assumed in this optimisation model and
therefore the total ownership cost per hour (Ctow(sh)) for loading equipment is

Ctow(sh) = x1 # cow(sh) (5)

where
x1 denotes to the number of shovel
Like as excavation equipment, depreciation component and interest component was considered
for unit ownership cost of haulage equipment (trucks). Based on Table 2 and Figure 3, the relation
between capital cost and truck capacity (x3) is as

Struck ( ) 14 914(x4) (6)

As mentioned before, homogenous fleet was assumed and total ownership cost per hour (Ctow(truck))
for truck is

Ctow(truck) = x3 # cow(truck) = x3 # ` D (truck) + I (truck)j (7)

Hourly operating costs are considered variable and are directly related to daily use. They should
be applied to daily use requirements, regardless of ownership period. The operating costs assume
the equipment is working a full operating hour under average operating conditions. Overhaul part,
overhaul labour, maintenance part, maintenance labour, fuel power (diesel fuel, gasoline, electric
power and natural gas), lubrication, tires and wear part are major component of operating costs.
Repair labour cost, diesel fuel, gasoline, natural gas, electricity and lubricant cost are assumed 29.97
/h, 1.078 /l, 0.927 /l, 7.663 /gig joule, 0.073 per kWh and 3.482 /l, respectively (CostMine,
2008).

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 306
 A NON-LINEAR PROGRAMMING MODEL FOR OPEN PIT MINE EQUIPMENT SELECTION

TABLE 2
Data of trucks, rear-dump and rigid frame costs (CostMine, 2008).

Trucks capacity (metric ton) Capital cost ($) Hourly operating costs ($/hour)
(US dollars, 2008) (US dollars, 2008)
31.7 472 000 50.65
36.3 619 000 65.32
36.3 617 000 65.27
40.8 637 000 67.64
59 1 017 000 91.75
54.4 798 000 86.87
52.6 829 000 87.67
63.5 851 000 98.79
77.1 1 212 000 119.27
77.1 1 191 000 122.01
90.7 1 128 000 139.45
90.7 1 329 000 137.21
99.8 1 305 000 166.77
154.2 1 882 000 180.01
136.1 2 052 000 206.46
154.2 2 434 000 241.91
190.5 2 774 000 276.53
181.4 2 115 000 296.53
217.7 3 051 000 360.27
254 5 102 000 430.72
290 3 550 000 444.96
327 4 103 000 527.41
327 5 634 000 563.6

FIG 3 - Relation between capital cost and truck capacity.

Total operating cost of shovels and trucks were illustrated in Tables 1 and 2. The relation between
shovel operating cost and bucket capacity (x2) is expressed as
sh
S ^ h 25.281 # ^ x2h (8)

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 307
A AGHAJANI BAZZAZI, M OSANLOO AND B KARIMI

where
S’ sh is the shovel operating cost
x2 is shovel capacity in cubic metres (Figure 4)

FIG 4 - Relation between operating cost and shovel bucket capacity.

The relation between number of shovel (x1), shovel capacity (x2) and operating cost is expressed
op(sh) ' (sh)
Ct = x1 # S x1 # 25.281 # x2 (9)

In similar manner, the relation between operating cost and truck capacity (x4) is expressed as

S'(truck)^ h 1.6043 # ^ x 4h (10)

where
S’ truck is the truck operating cost
x4 is truck capacity in metric ton (Figure 5)

FIG 5 - Relation between operating cost and truck capacity.

Therefore, capital cost for trucks fleet in open pit mines can be calculated by
op(truck) '(truck)
Ct = x3 # S x3 # 1.6043 # x4 (11)

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 308
 A NON-LINEAR PROGRAMMING MODEL FOR OPEN PIT MINE EQUIPMENT SELECTION

Consequently, objective function of this problem is expressed as

(loading - haulagee uipment) loading - haulagee uipment h
= min ^C +C (12)
op(shovel) op(truck)
Ctow(shovel) + Ct ow(truck)
+ Ct + Ct

x1 # c ow(sh) + x1 # S'(sh) + x3 # c ow(truck) + x3 # S'(truck)

sh sh sh
S ^r + i + ph^^ + 1h /2h o x # S'(sh)
x1 # e S - R + + 1
# H #H
truck truck
+ x3 # e S - Rtruck + S ^r + i + ph^^ + 1h /2h o x # S'(truck)
+ 3
#H #H

1686.8^ x2h2 + 192 273^ x2h - R sh 1686.8^ x2h2 + 192 273^ x2h^r + i + ph^^ + 1h /2h o
x1 # e +
#H #H
14 914^ x4h - Rtruck 14 914^ x4h^r + i + ph^^ + 1h /2h o + x # 1.6043 # ^ x4h
+ x1 # 25.281 # ^ x2h + x3 # e + 3
#H #H

Constraints
Shovel fleet annual productivity constraint
It is obvious that shovel fleet annual productivity should be determined rather than the rate of mine
annual production. Therefore, shovel productivity constraint should be expressed as

e# 1# 2 #^ 00/T h # S # # # Bf # # Hs # c (13)

where
x1 is number of shovels
x2 is shovel bucket capacity in cubic metres
Tc is shovel cycle time in seconds
SF is a swing factor – a correction for shovel cycle time to allow for variation in swing arc that
will be 1.0 if due allowance has been included in Tc otherwise SF for shovels can be estimated
from Table 3

TABLE 3
Correction factors for shovel cycle times (Osanloo, 2007).

Swing arc-degrees 45 50 60 90 100 130 180

Swing factor 1.26 1.16 1.07 1.00 0.88 0.77 0.7

A is available hours – the part of scheduled hours when the shovel is mechanically and
electrically ready to operate
U is utilised hours or operating hours – the part of the available hours that the shovel is actually
operating at expected productivity
Bf is bucket factor (ff/sw) – a random variable that combines the effects of in situ density, swell
and volumetric fill factor
ff is bucket fill factor – a variable that typically, for full bucket loads, can be expected to have
low dispersion (coefficient of variation of 1 or less) considered volumetrically
sw is swell factor a variable dependent on the inherent structural and geotechnical properties
PF is a propel factor ( 1) to allow for non-productive movement of the shovel to follow the
working face in the course of normal loading operations

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 309
A AGHAJANI BAZZAZI, M OSANLOO AND B KARIMI

Compared with rope shovels, hydraulic shovels have less reach so move more often but propel at
higher speed and do not require assistance with a trailing cable or cable bridge so a factor of 0.85 is
considered reasonable for all excavation equipment in open pit mines. Hs is scheduled hours – the
annual hours the shovel is expected to operate at expected productivity  is specific gravity and Pe is
annual productivity.
Original equipment manufacturers and equipment dealers generally provide average cycle times
for loading equipment items. Empirical values are tabulated for a range of digging conditions
(Table 4). Typically four levels of digging difficulty are specified in empirical terms viz, easy, medium,
hard and very hard. Consequently, shovel cycle time is dependent to shovel capacity(x2). Relation
between shovel cycle time and shovel bucket capacity can be expressed as (Figure 6)
0
c 0 Easy digging (14)
0
c 0 Medium digging

0
c 0 medium - hard digging
0
c 0 Hard digging

TABLE 4
Loading shovel cycle time (sec).

Shovel capacity (metric ton) E M M-H H

3 18 23 29 33
4 20 25 30 34
5 21 26 30 34
5.5 21 26 31 35
6 22 27 32 36
8 23 28 32 37
9 24 29 33 38
11.5 26 30 35 40
15 27 32 37 42
19 29 34 40 45
35 30 36 29 33

E Easy digging, loose, free-running material; eg sand, small gravel

M Medium digging, partially consolidated material; eg, clayey gravel, packed earth, clay, anthracite, etc
M-H Medium-hard digging; eg, well blasted limestone, heavy wet clay, weaker ores, gravel with large boulders, etc
H Hard digging- materials that require heavy blasting and tough plastic clays; eg granite, strong limestone, Taconite, strong ores, etc

Fleet matching constraint

Fleet matching is generally treated deterministically in terms of expected (mean) performance values
for both loading equipment and trucks. The perfect match point is hypothetical. It is where truck trip
and loading times have an exact integer relationship. Based on literature and experts viewpoint,
shovel bucket capacity should be three to six times of truck capacity and therefore this constraint
should be expressed as
3 # c # x2 # x4 # 6 # c # x2 (15)

Moreover, the relation between numbers of shovels (x1) and trucks numbers is expressed as
4 # x1 # x3 # 8 # x1 (16)

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 310
 A NON-LINEAR PROGRAMMING MODEL FOR OPEN PIT MINE EQUIPMENT SELECTION

FIG 6 - Relation between shovel capacity and shovel cycle time.

Number of trucks per shovel constraint

Trucks annual productivity should be determined rather than the rate of mine annual production.
Load and haul equipment selected for open pit mines, in fact for all mines, must satisfy the specific
needs of downstream facilities including practically constant production rate of mineral product.
Each truck has a productivity of

VT PL # 60 x4 # 60
(17)
TT TT
where
VT productivity per truck tonnes per hour (t/h)
PL or x4 truck payload in tonnes
TT truck trip time – minutes
Truck trip times represent in terms of four time components
1. manoeuvre-and-spot (TS) for loading,
2. manoeuvre-and-dump (TD),
3. loading (TL), and
4. travel (TV).
Accumulation of the four components is commonly termed ‘truck trip time’.
Manoeuvre and spot time for large mining trucks are suggested as ‘Usually between 0.4 to 0.7
minutes’ (Hays, 1990) or ‘Typically between 0.6 to 0.8 minutes’ (Caterpillar, 2004). As TS times are
a relatively small proportion of truck trip time, certainly
10 per cent, errors in estimating this time
component are of small effect. Adoption of 0.7 or 0.75 minutes will be generally acceptable.
TD time consists of two parts
1. Raising and lowering the body, generally ranging from 0.3 to 0.7 minutes.
2. Manoeuvring time. A typical range of TD times for rear dumping mining trucks is 1.0 to 1.2 minutes
(Caterpillar, 2004).
Truck-loading time can be expressed in terms of bucket cycle times as

From truck perspective – TL TF + (n-1). TC

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 311
A AGHAJANI BAZZAZI, M OSANLOO AND B KARIMI

where
n number of bucket cycles
TF time for dumping first bucket load
The relation between travel time (TV) and haul distance (DH) can be expressed as (subscript
‘L’ ‘loaded’ subscript ‘E’ ‘empty’)

TVL DHL (18)

SAVL DHL + DHE
& TV
TVE DHE SAV
SAVL

where
TV travel time
DHL haul distance loaded
DHE haul distance empty
SAV average speed
If the shovel does not have experience of waiting time, simple algebraic manipulation between
numbers of truck per shovel yields
x4 Tc T T T
x3 $ TT TS + TL + TV + TD c # x2 # 60 + S + V + D (19)
x1 TS + TL TS + TL x4 Tc T
c # x2 # 60 + S

Trucks fleet annual productivity constraint

As mentioned before, productivity per truck tonnes per hour (t/h) can be calculated by Equation 17.
In similar manner with Hs, Ht is scheduled hours, ie the annual hours the truck is expected to operate
at expected productivity. It is notable that trucks fleet annual productivity should be rather than the
rate of mine annual production. Therefore, trucks productivity constraint should be expressed as
P # # #^ 0 Th # # # # HT (
(20)
T ^ DHL + DHE h
P # # #e 0 e # c + TL + + TD oo # # # # HT
c# 0

Bench width constraint

Based on original equipment manufacturers catalogue, the relation between truck capacity and
maximum truck width can be found. In open pit mines, the minimum bench width must be four
times than truck width (Osanloo, 2007). Therefore, the relation between truck width and bench
width is expressed as (Figure 7 and Table 5)
4 # 1.526x40.305 # b
(21)

where
Wb is bench width

CASE STUDY
Anomaly two of Gol-e-Gohar iron ore mine of Iran contained 145 Mt material that included ore,
waste, over burden of 42, 43 and 60 Mt respectively. Annual production rate in this anomaly is
10 Mt and total over burden must be removed during six years. After over burden removal, 5 Mt
ore will be extracted with stripping ratio of 1 1. The mine’s life is 14 years. Hard digging condition
was considered for Tc calculation. Swell factor (sw) of haematite and magnetite is equal to 0.85
and bucket fill factor was assumed to 0.8 therefore bucket factor (Bf) is equal to 0.941 (0.8/0.85).
Scheduled hours for shovel were assumed to 3600 hours (two six-hour shifts in 300 days). A and

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 312
 A NON-LINEAR PROGRAMMING MODEL FOR OPEN PIT MINE EQUIPMENT SELECTION

FIG 7 - Relation between truck capacity and truck width.

TABLE 5
Relation between truck capacity and truck width.

Type Company Capacity Operating width

797B CAT 363T 9.154 m
793F CAT 250T 8.295 m
797F CAT 400T 9.116 m
785C CAT 150T 6.640 m
789C CAT 177T 7.674 m
793F CAT 250T 8.295 m
793D CAT 218T 7.680 m
EH750-3 HITACHI 42.5T 4.840 m
EH1100-3 HITACHI 72.3T 4.980 m
EH1700-3 HITACHI 100T 6.250 m
3500ACII HITACHI 185T 7.850 m
EH5000ACII HITACHI 314T 8.670 m
T282B LIEBHERR 363T 9.52 m
HD325-7R KOMATSU 36.5T 4.760 m
HD405-7R KOMATSU 41T 4.760 m
HD465-7R KOMATSU 55T 5.395 m
HD605-7R KOMATSU 63T 5.395 m
HD785-7 KOMATSU 91T 6.485 m
HD1500-5 KOMATSU 149T 6.62 m
730E KOMATSU 183.7T 7.54 m
830E KOMATSU 222.3T 7.32 m
930E KOMATSU 290.9T 9.09 m
TR100 TEREX 100T 5.9 m
MT6300AC TEREX 400T 9.70 m
MT4400AC TEREX 240T 7.97 m
MT3700AC TEREX 186T 7.30 m
MT5500AC TEREX 326T 9.154 m
MT3300AC TEREX 136T 8.295 m

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 313
A AGHAJANI BAZZAZI, M OSANLOO AND B KARIMI

U were assumed to 0.9 and 0.85, respectively. 0.85 is considered for PF factor and specific gravity
is equal to 3.2 ton per cubic metres. Ts and TD was assumed to 0.75 and one minute. Average haul
distance between pit and crushing plant was assumed 2 km and average speed of trucks were
considered to 30 k/h and therefore eight minutes was calculated for travel time (T V). Scheduled
hours for trucks (H T) were assumed as same as shovels scheduled time. Eighteen metre bench
widths were calculated for this mine. Typical life of mining equipment was illustrated in
Table 6. As mentioned before, annual scheduled hours for shovel is 3600 hours in this mine
and also total scheduled hours for this mine is about 50 400 hours. The life of equipment that
will be used in this mine is equal to mine’s life therefore replacement time of equipment was
not considered in this optimisation problem.
TABLE 6
Typical operating life of mining equipment (Runge, 1998).

Number of hours
Item
Poor condition Average condition Good condition
Dozer 18 000 25 000 35 000
Grader 20 000 30 000 50 000
Large front-end-loader 20 000 30 000 45 000
Hydraulic excavator (small) 15 000 22 500 30 000
Hydraulic excavator (large) 20 000 30 000 45 000
Scraper 12 000 16 000 20 000
Truck (50 to 100 t) 20 000 30 000 42 000
Truck (large) 30 000 45 000 60 000
Rope shovel 60 000 80 000 100 000
Walking dragline 60 000 1 00 000 15 000

Salvage value of shovels and trucks equipment is assumed to 20 per cent of capital cost, interest
rate on capital of excavation and trucks is equal to 12 per cent, insurance rate of equipment was
assumed to two per cent and equipment taxes are considered to five per cent in objective function.
The final form of loading haulage equipment selection problem in Anomaly two of Gol-e-Gohar iron
mine is summarised in Equation 22.
0.8 # ^1686.8^ x2h2 + 192 273^ x2hh ^1686.8^ x2h2 + 192 273^ x2hh # ^0.12 + 0.02 + 0.05h # ^^14 + 1h /2h
min e x1 # e + oo
30 000 30 000
0.8 # ^14 914^ x4hh 14 914^ x4h # ^0.12 + 0.02 + 0.05h # ^^14 + 1h /2h
+ x1 # 25.281 # ^ x2h + x3 # c + m + x3 # 1.6043 # ^ x4h
30 000 30 000
subject to
10 000 000 # x1 # x2 # ^3600/26.92x20.145h # 1 # 0.9 # 0.85 # 0.941 # 0.85 # 3600 # 3.2

3 # 3.2 # x2 # x4 # 6 # 3.2 # x2
(22)
4 # x1 # x3 # 8 # x1
x4 # Tc + 0.75 + 8 + 1
x3 $ c # x2 60
x1 x4 # Tc + 0.75
c # x2 60
x4 26.92x20.145
10 000 000 # x3 # x4 # e60/ e # + 0.75 + 8 + 1 oo # 0.9 # 0.85 # 0.941 # 3600
3.2 # x2 60
4 # 1.526x40.305 # 18

For solving fitness function, MATLAB R2008a software was used and after running this optimisation
problem, 5.6 shovels with 2 m3 and 44.7 trucks with 22 metric ton capacity were obtained. It is
obvious that number of shovels and truck must be round up as six shovels with two cubic metres
capacity and 45 trucks with 22 metric ton capacity. This was proposed for this open pit mine with
approximately 3200 per hour cost.

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 314
 A NON-LINEAR PROGRAMMING MODEL FOR OPEN PIT MINE EQUIPMENT SELECTION

CONCLUSION
Equipment selection is one of the important aspects of mine planning and although the optimum
equipment selection for a mine is a complex decision, it can be broken down into a series of relatively
well-defined variables that are directly related to the general mine conditions.
In this paper we extend non-linear programming (NLP) with continuous variables to an open pit
mine loading-haulage equipment selection problem. The model integrates capital and operating cost
in objective function, shovel and truck annual productivity constraint, fleet matching constraint and
number of truck per shovel constraint. This model was applied in Gol-e-Gohar mine and the result
was verified. Incorporation of annual production uncertainty and other goal function in this type of
optimisation problem are suggested for future research.

REFERENCES
gha ani Ba a i, A, Osanloo, M and Soltanmohammadi, H, 2008. loading-haulage equipment selection in
open pit mines based on Fuzzy-Topsis method, Gospodarka Suro cami Mineralnymi-Mineral Resources
Management, 24(3) 87-102.
gha ani Ba a i, A, Osanloo, M and Karimi, B, 2009. Optimal open pit mining equipment selection using
fuzzy multiple attribute decision making approach, Archi es of Mining Sciences, 54(2) 301-320.
Bandopadhyay, S, 1987. Partial ranking of primary stripping equipment in surface mine planning,
International Journal of Mining, Reclamation and En ironment, 1(1) 55-59.
Bandopadhyay, S and Venkatasubramanian, P, 1987. Expert systems as decision aid in surface mine
equipment selection, International Journal of Mining, Reclamation and En ironment, 1(2) 159-165.
Bas etin, A, 2004. An application of the analytic hierarchy process in equipment selection at Orhaneli open
pit coal mine, Transactions of the Institutions of Mining and Metallurgy, Mining Industry, 113 192-199.
Bo orgebrahimi, E, Hall, R, Blackwell, G and Scoble, M, 2003. Economies of scale in surface mining A
case study of the impact of haul truck size on maintenance cost, SME Annual Meeting, Cincinnati, Ohio,
Preprint 03-020.
Caterpillar, 2004. Caterpillar Performance Handbook, edition 35 (Caterpillar Inc Peoria).
Cebesoy, T M, Gozen and ahsi, S, 1995. A systematic decision making model for optimum surface mining
equipment, in Proceedings Fourth International Symposium on Mine Planning and E uipment Selection,
pp 369-377 (Balkema Rotterdam).
CostMine, 2008, Surface Mining E uipment (CostMine Inc Washington).
Denby, B, Clarke, M P and Schofield, D, 1990. Decision making tools for surface mining equipment selection,
Mining Science and Technology, 10(3) 323-335.
Denby, B and Schofield, D, 1990. Application of expert systems in equipment selection for surface design,
International Journal of Mining, Reclamation and En ironment, 4(4) 165-171.
Ganguli, R and Bandopadhyay, S, 2002. Expert system for equipment selection, International Journal of
Mining, Reclamation and En ironment, 16(3) 163-170.
Gregory, B S, 2003. Excavator selection, in Proceedings 1 th International Symposium on Mine Planning
and E uipment Selection, pp 395-404 (The Australasian Institute of Mining and Metallurgy Melbourne).
aidar, A and Naoum, S, 1995. Selection of opencast mine equipment using knowledge base and genetic
algorithm, in Proceedings Fourth International Symposium on Mine Planning and E uipment Selection,
pp 409-413 (Balkema Rotterdam).
artman, H and Mutmanski, J M, 2002. Introductory Mining Engineering (John Wiley Sons, Inc New Jersey).
ays, R M, 1990. Surface mining, Trucks (ed B A Kennedy), pp 672-691 (Society for Mining, Metallurgy and
Exploration, Inc Littleton).
rebar, M J, 1990. Preliminary dragline selection for surface coal mining operation, in Proceedings Second
International Conference on Mine Planning and E uipment Selection, Calgary, pp 133-143.
aya ardane, A K W and Harris, F C, 1990. Further development of integer programming in earthwork
optimization, Journal of Construction engineering and management, 116(7) 18-34.
Maitra, S, Rao, P V, Sengupta, D and Rao, V, 1994. Optimal production scheduling using operations research
techniques in a large opencast iron mine, in Proceedings Third International Symposium on Mine Planning
and E uipment Selection, Istanbul, pp 87-91.
sanloo, M, 2007. Surface Mining Methods, pp 453-638 (in Persian).

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 315
A AGHAJANI BAZZAZI, M OSANLOO AND B KARIMI

ao, K, Reddy, V and Sathaiah, C, 1989. Equipment assessment with life-cycle costing (LCC) – A case study
for loader selection in underground coal mines, Minetech, 10(6) 12-18.
e ni henko, S and Valuev, A, 1994. Simulation of mining dynamics for middle- and short-term open pit
production planning, in Proceedings Third International Symposium on Mine Planning and E uipment
Selection, pp 93-97 (Balkema Rotterdam).
unge, I C, 1998. Mining Economics and Strategy, 295 p (Society of Mining, Metallurgical and Exploration,
Inc Littleton).
Samanta, B, Sarkar, B and Mukherjee, S, 2002. Selection of opencast mining equipment by a multi-criteria
decision-making process, Transactions of the Institutions of Mining and Metallurgy, Mining Technology,
111 A136-A142.
Singhal, R, 1986. Surface mining equipment selection and use are the keys to profits, orld Mining
E uipment, 10(1) 10-16.
Singhal, R and Fytas, K, 1986. Truck selection in surface mining, orld Mining E uipment, 10(1) 18-26.
in hun, L and oudi, , 2004. Application of genetic algorithm to optimize the number and size of equipment
of surface mine, in Proceedings 1 th International Symposium on Mine Planning and E uipment Selection,
pp 638-640 (Balkema Rotterdam).

MINE PLANNING AND EQUIPMENT SELECTION (MPES) CONFERENCE / FREMANTLE, WA, 1 - 3 DECEMBER 2010 316