JME

Journal of Mining & Environment,

Vol.2, No.2, 2011, 118-125.

A new algorithm for optimum open pit design: Floating cone method III
E. Elahi zeyni1, R. Kakaie2*, A. Yousefi3
1- Eastern Alborz Coal Company, Shahrood
2- Faculty of Mining Engineering & Geophysics, ShahroodUniversity of Technology, Shahrood
1- Eastern Alborz Coal Company, Shahrood
Received 4 Oct 2011; received in revised form 9 Mar 2012; accepted 5 Apr 2012
*Corresponding author: [email protected] (R. Kakaie)

Abstract
Ultimate limits of an open pit, which define its size and shape at the end of the mines life, is the pit with the
highest profit value. A number of algorithms such as floating or moving cone method, floating cone method
II and the corrected forms of this method, the Korobov algorithm and the corrected form of this method,
dynamic programming and the Lerchs and Grossmann algorithm based on graph theory have been developed
to find out the optimum final pit limits. Each of these methods has special advantages and disadvantages.
Among these methods, the floating cone method is the simplest and fastest technique to determine optimum
ultimate pit limits to which variable slope angle can be easily applied. In contrast, this method fails to find
out optimum final pit limits for all the cases. Therefore, other techniques such as floating cone method II and
the corrected forms of this method have been developed to overcome this shortcoming. Nevertheless, these
methods are not always able to yield the true optimum pit. To overcome this problem, in this paper a new
algorithm called floating cone method III has been introduced to determine optimum ultimate pit limits. The
results show that this method is able to produce good outcome.
Keywords: Open pit mining; Ultimate pit limit; Floating cone method; Floating cone method II; Floating
cone method III
1. Introduction
Open pit mining is an important general mining
method that mineral deposit will be mined via
pits. The shape of mining area at the end of
mining operation or final limits of a mine must be
designed before starting the operation. According
to the designed final pit limits, mining operational
parameters such as width, length and depth of
mined pit, opening track ways, location of waste
dump, stripping ration, mine life, minable ore
tonnage, waste tonnage and production scheduling
can also be determined [1].
Optimum pit limits are usually designed with the
use of the block models. Geological block model,
which presents the reserve as a combination of
numerous small blocks, is determined by inverse
distance or geostatistical methods. Then the
economical block model is calculated by applying

cost, price and other parameters to each block. In

this model ore blocks have positive values, waste
blocks have negative values and air blocks, and
the blocks over the surface topography have zero
values. Most of the optimum pit limits methods
use the economical block models to determine the
pit limits. The methodology is searching for a
combination of blocks with the maximum
economical value at current economical and
technical condition [2].
Floating or moving cone method [3], floating cone
II method [4], modified floating cone II methods
[5], dynamic programming [6], [7] and [8], the
Lerchs and Grossmann algorithm based on graph
theory [9], Korobov algorithm [10], corrected
form of the Korobov algorithm [11] and genetic
algorithm [12] are some of the several algorithms

Elahizeyni et al./ Journal of Mining & Environment, Vol.2, No.2, 2011

developed to determine optimum pit limits. Each

of these methods has special advantages and
disadvantages.
For
example,
dynamic
programming is just a 2D modeling method and
although the Lerchs and Grossmann algorithm can
always create an optimized and true limit of the
pit, the method is too complicate to be applied
easily.
Floating or moving cone algorithm is one of the
easiest and fastest algorithms for determining the
final pit limits. In addition, mining operational
restrictions on various slopes can be applied to
this method perfectly. Further, a few algorithms
such as floating cone II method and modified
floating cone II method are developed to
overcome the shortcoming of this method. In
despite of these corrections, these methods are not
able to obtain the true optimum limit of the pit.
Therefore it needs more corrections. This paper
presents a new edition of the algorithm, called
floating cone III method, in order to cover
shortcomings of previous methods.

1
2
3

1
-3
-4
-5

2
-3
7
-5

3
-3
-4
-5

4
-3
6
15

5
-3
-4
-5

6
-3
-4
-5

7
-3
-4
-5

8
-3
8
-5

2
-3
7
-5

3
-3
-4
-5

4
-3
6
15

5
-3
-4
-5

6
-3
-4
-5

7
-3
-4
-5

8
-3
8
-5

9
-3
-4
-5

Figure 2. Optimum pit limit by dynamic programming

method

Table 1. Cone value for example shown in Figure 1

Stage

Block
No

Block
value

Cone
value

Minable?

(2,2)

+7

-2

No

2
3
4

(2,4)
(2,8)
(3,4)

+6
+8
+15

-3
-1
-2

No
No
No

2-2- Floating cone II algorithm

Floating cone II algorithm was introduced and
presented by Wright in 1999 [4]. The
methodology is similar to the floating cone
approach except that first values of the cone of all
ore blocks are calculated in each level and the
cone with maximum value is removed from the
block model. Next cumulative pit value is
calculated and this process is carried out for
remaining ore blocks. Then all the extraction
cones of the block with highest cumulative pit
value are included as a member of the optimum
solution set. For the block model shown in Figure
1, when final dip of pit is 1:1, the result of this
algorithm is illustrated in Table 2 and Figure 3,
with the value of -3. A true optimum pit limit of
this model is shown Figure 2 with the value of +2.

2- An overview on floating cone algorithms

2-1- Floating cone method
This method, which was first described by
Carlson, Erickson, OBrain and Pana (1966),
works on an economical block model of the
deposit [3]. For each positive (ore) block, this
method involves constructing a cone with sides
oriented parallel to the pit slope angles, and then
determining the value of the cone by summing the
values of blocks enclosed within it. If the value of
the cone is positive, all blocks within the cone are
mined. This process starts from the uppermost
level and moves downward searching for positive
blocks. The process continues until no positive
cones remain in the block model. Although this
algorithm is simple and easy to understand, it is
not able to yield a true optimum pit limit. For
example, for the 2-D economical block model in
Figure 1, when final dip of pit is 1:1, floating cone
algorithm cannot produce true optimum pit limit,
as presented on Table 1. Nevertheless, by
applying dynamic programming method [6] to this
model an optimum pit limit with the value of 2
would be obtained (Figure 2).
1
2
3

1
-3
-4
-5

Table 2. Cumulative pit value for example shown in

Figure 1
L
e
v
e
l

2
3

1
2
3

9
-3
-4
-5

Block
No

Block
value

Cone
value

Cumulative
value

Minable
?

(2,8)
(2,2)
(2,4)
(3,4)

+8
+7
+6
+15

-1
-2
0
-2

-1
-3
-3
-2

Yes
No
No
Yes

1
-3
-4
-5

2
-3
7
-5

3
-3
-4
-5

4
-3
6
15

5
-3
-4
-5

6
-3
-4
-5

7
-3
-4
-5

8
-3
8
-5

9
-3
-4
-5

Figure 3. Optimum pit limit by floating cone II algorithm

Figure 1. 2-D economical block model

119

Elahizeyni et al./ Journal of Mining & Environment, Vol.2, No.2, 2011

3-2- Finding ore blocks from the first level of

economical block model to the other levels. If any
ore block is found in any level, other ore blocks
are considered from this level to the first level.
The aim of this stage is to check the effect of
levels on each other.
3-3- Constructing extraction cones for all ore
blocks, with regard to the technical restrictions,
then finding dependent and independent blocks.
3-4- Finding ineffective and independent blocks
which have no positive values. If positive values
are assigned against negative valued blocks, the
effect of such blocks on optimum pit limit will be
removed.
3-5- Finding ineffective and dependent blocks
which have no positive values. If positive values
are assigned against uncommon and negative
overlying blocks, the effect of such blocks on
optimum pit limit will be removed or decreased.
3-6- Finding effective and dependent blocks, these
are all remaining blocks after carrying out the
foregoing stages. Finding optimum pit limit is as
follows:
3-6-1- Identifying common blocks for each of
extraction cone and then calculating their weights,
The weight of each block is equal to the number
of cones enclosed within it.

2-3- Modified floating cone II, method 1

This algorithm is the same as the algorithm of
floating cone II method except that when the
highest cumulative pit value is positive, the blocks
are added as part of the optimum pit solution [5].
Therefore using this method, for the block model
shown in Figure 1, as illustrated in Table 2 3, no
blocks would be mined. Nevertheless, the true
optimum limit for this model is shown in Figure 2
with the value of +2.
2-4- Modified floating cone II, method 2
This method is a development of first method of
modified floating cone II algorithm [5]. In the
second method all levels are considered together.
In other words, the value of the cones of all ore
blocks are evaluated economically and the cone
with maximum value is assumed to be as part of
the pit limits and the cumulative pit value is
calculated. This process is then continued until no
positive block remains in the block model.
Finally, the block with positive and maximum
cumulative value and all other previous blocks are
included as a member of the optimum solution set.
As illustrated in Table 4 and Figure 4, this
algorithm yields a pit limit with the value of +1
for the block model shown in Figure 1.
3- Floating cone III
Although the floating cone II method and its
corrections overcomes some weaknesses of the
floating cone method, in some circumstances
these methods fail to yield a true optimum pit
limit and therefore a new development is needed.
In general, blocks in economical block models
could be divided into two groups of dependent
and independent blocks. The ore blocks which
have no common overlying block with other ore
blocks in their extraction cones are independent
otherwise they are classified as dependent. Each
of these groups with regard to the value of their
extraction cones are also classified as effective or
ineffective. Effective blocks have positive value
and ineffective blocks have negative value. The
optimum pit limit is determined according to the
cones on effective blocks. A flowchart of the
floating cone III method is shown in Figure 5.
Stages of the algorithm are as follows:
3-1- The algorithm is the same as the floating
cone algorithm except that after extraction of each
mining cones, search for other limits would be
continued from the first level for remaining
blocks. The aim of this stage is finding
independent effective blocks in economical
model.

Table 3. Cumulative pit value by floating cone II,

method 1
L
e
v
e
l

2
3

Block
No

Block
value

Cone
value

Cumulative
value

Minable
?

(2,8)
(2,2)
(2,4)
(3,4)

+8
+7
+6
+15

-1
-2
0
-2

-1
-3
-3
-2

No
No
No
No

Table 4: Cumulative pit value by floating cone II,

method 2
Stage

Block
No

Block
value

Cone
value

Cumulative
value

Minab
le?

1
2
3

(2,8)
(2,2)
(3,4)

+8
+7
+15

-1
-2
+4

-1
-3
+1

Yes
Yes
Yes

1
2
3

1
-3
-4
-5

2
-3
7
-5

3
-3
-4
-5

4
-3
6
15

5
-3
-4
-5

6
-3
-4
-5

7
-3
-4
-5

8
-3
8
-5

9
-3
-4
-5

Figure 4. Optimum pit limit by floating cone II, method 2

120

Elahizeyni et al./ Journal of Mining & Environment, Vol.2, No.2, 2011

Start

Continue for remaining blocks

with original value

Take first level

Include this cone as part of the

pit
Yes

Is there any independent

and effective ore block?
Take next level

No
All levels?

No

Yes
Take first level with
original value

Yes

Is there any independent

and ineffective ore block?

Construct cone and allocate positive

value against negative value

No

Yes

Is there any dependent and

ineffective ore block?

Construct cone and allocate positive value

against negative value for uncommon block

No

Is there any positive

block?
No

Yes
Construct cone and calculate importance of
common block, value of cone and final importance

Sort ore blocks in ascending order on their

importance and cone value

Remove cones in ascending order and calculate

cumulative values

Is maximum cumulative
value positive?
No
Take next level

No

All levels?
Yes
End

Figure 5. Flowchart of Floating Cone III algorithm

121

Yes

Include all the cones as part of the pit for

the maximum cumulative value

Elahizeyni et al./ Journal of Mining & Environment, Vol.2, No.2, 2011

3-6-2- Calculating the weight of mining cones. It

is the sum of weight of common blocks of the
cone.
3-6-3- Calculating the cone values for each of
mining cone.
3-6-4- Calculating the final importance for each of
mining cone.Final importance is the ratio of the
weight of cone to the absolute value of its cone
value.
3-6-5- Sorting mineral blocks on final importance,
then on value of mining cones in descending order
and supposing extraction of first ore block.
3-6-6- Supposing extraction of other blocks would
be continued according to fist block extraction. At
this stage the value of other cones will be
evaluated and the mining cone with maximum
value will be extracted, a supposed extraction
offcourse. The cumulative value of every
supposed extraction must be calculated now.
3-6-7- Finding the maximum and positive
cumulative value and determining optimum pit
limit by including blocks from fist mining cone to
this cone. Then we search for another limit from
next levels by using original block model. If the
maximum cumulative value is not positive, this
means that there is no optimum limit to this level
and the search will be continued from next level.
The floating cone III algorithm can best be
explained by a simple example applied to an
economical model shown in Figure 1. As shown
below, this algorithm produces a final pit with the
value of +2 (Figure 13) which is a true optimum
pit limit.
a) The first level containing ore blocks is the
second level and the algorithm applied for this
level is as follows:
First stage: finding effective ore blocks. As can be
seen in Table 1, since the extraction cone of all
the ore blocks are negative, there is no
independent effective block in the model.
Second stage: Removing independent and
ineffective ore blocks from model, as shown in
Figure 6.
1
2

1
-3
-4

2
-3
7

3
-3
-4

4
-3
6

5
-3
-4

6
-3
-4

7
0
-4

8
0
0

block here and the initial and final importance of

this cone is 0, as is shown in Figure 8.
1
2

1
0
-4

2
0
1

3
-3
-4

4
0
0

5
0
-4

6
-3
-4

7
0
-4

8
0
0

9
-1
-4

Figure 7Decreasing or removing the effect of dependent

and ineffective blocks

1
2

1
0
0

2
0
0

3
0
0

4
0
0

5
0
0

6
0
0

7
0
0

8
0
0

9
0
0

Figure 8. Determining importance of common blocks

Fifth stage: Supposing the extraction of only

remaining ore block which its value of extraction
cone and its cumulative value of both are -2. Since
the value is less than zero, this cone is not
included as part of pit limit. In other words, there
is no optimum limit to the second level. Hence the
algorithm will be continued for the economical
block with its original value as follows:
b) The second level with ore blocks is the third
level and the stages of the algorithm for this level
are as follows:
Sixth stage: finding effective ore blocks. As it can
be seen from Table 1, there is no independent
effective block in the model.
Seventh stage: Removing independent and
ineffective ore blocks, as illustrated in Figure 9.
1
2
3

1
-3
-4
-5

2
-3
7
-5

3
-3
-4
-5

4
-3
6
15

5
-3
-4
-5

6
-3
-4
-5

7
0
-4
-5

8
0
0
-5

9
-1
-4
-5

Figure 9. Removing independent and ineffective ore

blocks

Eighth stage: decreasing or removing the effect of

dependent and ineffective blocks, as shown in
Figure 10.
1
2
3

1
0
-4
-5

2
-3
4
-5

3
-3
0
-5

4
-3
6
4

5
-3
0
-5

6
0
-4
-5

7
0
-4
-5

8
0
0
-5

9
-1
-4
-5

Figure 10. Decreasing or removing the effect of dependent

and ineffective blocks

9
-1
-4

Ninth stage: Determining the importance of

common blocks, as shown in Figure 11.

Figure 6. Removing independent and ineffective ore

blocks

Third stage: decreasing or removing the effect of

dependent and ineffective blocks, as illustrated in
Figure 7.
Forth stage: Determining the importance of
common blocks. According to Figure 7, only one
ore block remains. Therefore there is no common

1
2
3

1
0
0
0

2
2
0
0

3
3
0
0

4
2
2
0

5
2
0
0

6
0
0
0

7
0
0
0

8
0
0
0

9
0
0
0

Figure 11. Determining importance of common blocks

122

Elahizeyni et al./ Journal of Mining & Environment, Vol.2, No.2, 2011

Tenth stage: Descending sort of ore blocks on

their final importance of mining cones, and then
on the value of cones, as shown in Table 5.

9
-2
-4
-6

Table 5. Descending sort of the ore blocks

8
-2
-4
-6

7
-2
-4
17

6
-2
-4
-6

5
-2
-4
-6

4
-2
-4
16

3
-2
-4
14

2
-2
-4
-6

1
-2
-4
-6

1
2
3

Figure 14. An economical block model

Ref

Block
No

Block
value

Final
importance

Cone
value

1
2
3

(3,4)
(2,4)
(2,2)

+4
+6
+4

5
3
2.5

-2
-3
-2

Table 7. Stages of floating cone II and its

modification algorithm

L
e
Block
v No
e
l

Eleventh stage: Supposed extraction of ore blocks

as illustrated in Figure 12.

Block
value

Cone
value

Cumulative
value

Minable
?

+17
+16
+14

-5
-2
+8

-5
-7
+1

Yes
Yes
Yes

(3,7)
1
2
3

1
-3
-4
-5

7
-5

-5

-5

-4
-5

7
0
-4
-5

8
0
0
-5

3 (3,4)

9
-1
-4
-5

(3,3)

9
-2
-4
-6

Figure 12. First supposed limits

Twelfth stage: Finding maximum value of

remaining blocks from sorted list and calculating
their cumulative value (Table 6).

Block
No

Block
value

Cone
value

Cumulative
value

Minable
?

1
2

(3,4)
(2,2)

+3
+7

-3
+4

-3
+1

Yes
Yes

9
0
-4
-6

8
0
0
-5

7
0
-4
-5

6
-3
-4
-5

5
-3
-4
-5

4
-3
6
15

3
-3
-4
-5

2
-3
7
-5

1
-3
-4
-5

6
-2
-4
-6

5
-2
-4
-6

4
-2
-4
16

3
-2
-4
14

2
-2
-4
-6

1
-2
-4
-6

1
2
3

8
0
-1
-6

7
0
0
0

6
-2
0
-6

5
-2
0
-6

4
-2
-4
12

3
-2
-4
8

2
-2
0
-6

1
0
-4
-6

1
2
3

Figure 16-a. Dependent ineffective blocks removalfloating cone III algorithm

Thirteenth stage: With regard to Table 6,

maximum cumulative value is positive. Therefore
related mining cones is included as part of the pit
and optimum limit is obtained with the value of
+2 (Figure 13). The result of this algorithm is the
same as the Dynamic programming method.
9
-1
-4
-5

7
-2
-4
17

Figure 15. Optimum pit limit by of floating cone II and its

modification algorithm

Table 6. Calculation of cumulative values

Ref

8
-2
-4
-6

9
0
0
0

8
0
0
0

7
0
0
0

6
2
0
0

5
2
0
0

4
2
2
0

3
2
2
0

2
2
0
0

1
0
0
0

1
2
3

Figure 16-b. Determining the importance of dependent

and ineffective common blocks - floating cone III
algorithm

9
-2
-4
-6

1
2
3

8
-2
-4
-6

7
-2
-4
17

6
-2
-4
-6

5
-2
-4
-6

4
-2
-4
16

3
-2
-4
14

2
-2
-4
-6

1
-2
-4
-6

1
2
3

Figure 13.Final Optimum pit limit by floating Cone III

Figure 16-c. Optimum pit limit by floating cone III

algorithm

Example 2
In order to show the ability of the floating cone III
algorithm, another simple example as shown in
Figure 14, When final dip of pit is 1:1, is
employed. Floating cone II algorithm and its
modification produce a pit with value of +1 as
illustrated in Figure 15, whereas floating cone III
method as shown in Table 8 and Figure 16-c
creates a true optimum pit with value of +2 as the
same as the dynamic programming technique.

3-7- Case study

In this section, floating cone algorithms are
applied for a real data of a gold mine, located at
35 kilometers north-east of Sweden. First of all,
an economical block model of this mine has been
created using Pitwin32 software. This software
with using grade block model and technical and
economical parameters such as cut-off grade,
dimension of blocks, ore and waste density, price,

123

Elahizeyni et al./ Journal of Mining & Environment, Vol.2, No.2, 2011

cost and etc creates an economical block model of

deposit.
The deposit is divided into 15m (east-west) x 10m
(north-south) x 5m (vertical) blocks and the lock
numbers bin the east west, north south and
vertical directions are 101, 82 and 36 respectively.
Each block is assigned the estimated (kriged)
recoverable tonnage of ore above a cut-off grade
and the estimated (kriged) average grade of this
tonnage. Table 9 shows the overall results of this
case study with the use of floating cone, floating
cone II and its modifications methods. In addition,
for implementation of the floating cone III method
a C++ code was developed by using Visual C++
programming language. The result of this
algorithm for this case is also shown in Table 9. It
can be concluded from this table that compared

with other floating cone methods, floating cone III

produces a final pit limit with the highest value.
4- Summary
Although floating cone II algorithm and its
modifications overcome some of the shortcomings
of the floating cone method, for examples shown
in this paper they produce a pit with less value and
fail to determine true optimum pit limits. Since
these methods do not take into account the effect
of independent and dependent block to each other.
On the other hand, the floating cone III method
take into consideration this shortcoming and
always create a pit with positive and high value.
The algorithm is straightforward and using
different pit slopes in different parts of the
orebody is very simple.

Table 8. Descending sorting of dependent effective blocks - floating cone III algorithm
Ref
Block No
Block value
Final
Cone value
Cumulative
Minable?
importance
value of cone
1
(3,4)
+12
2.33
-6
-6
Yes
2
(3,3)
+8
1.5
+8
+2
Yes

Method

Table 9. Overall results by different methods

Number of blocks
Value (*10000)

Floating cone
Floating cone II
Floating cone II- First modification
Floating cone II- Second
modification
Floating cone III
Lerchs& Grossmann

Pit
10053
12351
12351
13732

Ore
4394
4932
4932
5341

Waste
5660
7419
7419
8391

Ore
76895.5
82544.3
82544.3
85182.1

Waste
-11597.4
-15437.8
-15437.8
-17374.7

Net
65298.1
67106.5
67106.5
67807.4

15027
15030

5584
5405

9443
9625

88138.1
89853.7

-19553
-20324.9

68585.1
69528.8

[5] Khalokakaie, R. 2006; Optimum open pit design

with modified moving cone II methods, Journal of
engineering in Tehran university, Vol. 4, No. 3 p.p
297-307 (in Persian).

References
[1] Khalokakaie, R., Dowd, P. A. and Fowell, R. J.,
2000, Lerchs-Grossmann algorithm with variable
slope angles, Trans. Instn Min. Metall. (Sect. A: Min.
industry), No. 109, p.p. A77-A85.

[6]Koenigsberg E.; 1982; The optimum contours of an

open pit mine: an application of dynamic
programming, Proceedings of the 17th Symposium on
the application of computers and operations research in
the mineral industries (APCOM), (New York: AIME),
p.p. 247-287.

[2] Khalokakaie, R.; Dowd, P. A. and Fowell, R. J.;

2000; A windows program for optimal open pit design
with variable slope angles, International Journal of
Surface Mining, Reclamation and Environment, 14,
p.p. 261-275.

[7] Wilke, F. L. and Wright, E. A.; 1984;

Determining the optimal ultimate pit design for hard
rock open pit mines using dynamic programming,
Erzmetall, No. 37, p.p. 139-144.

[3] Carlson, T. R.; Erickson, J. D.; OBrain D. T. and

Pana, M. T.; 1966; Computer techniques in mine
planning, Mining Engineering, Vol. 18, No. 5, p.p.
53-56.

[8] Yamatomi, J.; Mogi, G.; Akaike, A. and

Yamaguchi, U.; 1995; Selective extraction dynamic
cone algorithm for three-dimensional open pit
designs, Proceedings of the 25th Symposium on the
application of computers and operations research in the
mineral industries (APCOM), Brisbane, p.p. 267-274.

[4] Wright, E. A.; 1999; MOVING CONE II - A

Simple Algorithm for Optimum Pit Limits Design,
Proceedings of the 28th Symposium on the application
of computers and operations research in the mineral
industries (APCOM), (Colorado USA), p.p. 367-374.

124

Elahizeyni et al./ Journal of Mining & Environment, Vol.2, No.2, 2011

[9] Lerchs, H. and Grossmann, I. F.; 1965;

Optimumm design of open pit Mine; CIM Bulletin,
No. 58, pp. 47 54.
[10] David, M.; Dowd, P. A. and Korobov, S.; 1974;
Forecasting departure from planning in open pit
design and grade control, Proceedings of the 12th
Symposium on the application of computers and
operations research in the mineral industries
(APCOM), (Golden, Colo: Colorado School of Mines),
Vol. 2, p.p. F131-F142.
[11] Dowd, P. A. and Onur, A. H.; 1993; Open pit
optimization - part 1: optimal open pit design, Trans.
Instn Min. Metall. (Sect. A: Min. industry), No. 102,
p.p. A95-A104.
[12] Denby, B. and Schofield, D.; 1994; Open-pit
design and scheduling by use of genetic algorithms,
Trans. Instn Min. Metall. (Sect. A: Min. industry), No.
103, p. p. A21-A26.

125