International Journal of Mining Science and Technology 25 (2015) 225–230

Contents lists available at ScienceDirect

International Journal of Mining Science and Technology

journal homepage: www.elsevier.com/locate/ijmst

Optimum mining method selection using fuzzy analytical hierarchy

process–Qapiliq salt mine, Iran
Karimnia Hamed a,⇑, Bagloo Heydar b
a
Department of Mining Engineering, Petroleum and Geophysics, Shahrood University of Technology, Shahrood, Iran
b
Department of Mining & Metallurgy Engineering, Amirkabir University of Technology, Tehran, Iran

a r t i c l e i n f o a b s t r a c t

Article history: Mining method selection is the first and the most critical problem in mine design and depends on some
Received 13 July 2014 parameters such as geotechnical and geological features and economic and geographic factors. In this
Received in revised form 17 August 2014 paper, the factors affecting mining method selection are determined. These factors include shape, thick-
Accepted 21 October 2014
ness, depth, slope, RMR and RSS of the orebody, RMR and RSS of the hanging wall and footwall. Then, the
Available online 30 March 2015
priorities of these factors are calculated. In order to calculate the priorities of factors and select the best
mining method for Qapiliq salt mine, Iran, based on these priorities, fuzzy analytical hierarchy process
Keywords:
(AHP) technique is used. For this purpose, a questionnaire was prepared and was given to the associated
Mining method selection
Fuzzy AHP
experts. Finally, after a comparison carried out based on the effective factors, between the four mining
Multiple criteria decision making methods including area mining, room and pillar, cut and fill and stope and pillar methods, the stope
Salt mine and pillar mining method was selected as the most suitable method to this mine.
Ó 2015 Published by Elsevier B.V. on behalf of China University of Mining & Technology.

1. Introduction accurate analysis of available information. Mining method selec-

tion is often affected by the following factors:
Mining method selection is one of the most sensitive stages of
mine design. The factors such as deposit features, economic vari- (1) Physical and mechanical characteristics of deposit such as
ables and geographic characteristics are investigated in order to local geological conditions, the strengths of the hanging-
select the best mining method considering the main purposes and footwall, ore thickness, general shape of the deposit,
including the maximization of the product level and profit, secur- slope, overburden thickness, grade distribution, and quality.
ing the safety of the workers, full extracting of the mineral, and The main geological conditions include shear strength of the
minimization the environment impact. The important factors for intact rock, natural fractures, shear strength of the in conti-
mining method selection include geological and mineralogical nuities, the orientation, length, spacing and locality of the
factors, geometrical features of deposit, safety, environmental geological structures, in situ stresses and hydrological
and economic factors and local considerations. Mining method conditions.
replacement causes cost and time overruns, hence, the best mining (2) Economic factors such as capital cost, operational costs,
method should be selected as accurate as possible at the mine extractability of the ore, grade and value of the mineral.
design stage. Every deposit has its unique features, and engineer- (3) Technical factors such as mine rehabilitation, method flex-
ing judgments have significant effects on the decision making ibility, equipment and extraction amount.
about mining method. Therefore, it is obvious that just an (4) Exploitation factors such as annual exploitation, equipment,
experienced engineer, who has adequate skills related to different efficiency and environmental aspects.
kinds of mining methods, can make the best decision. However,
experiences and judgments of engineering still present the neces- However, each criterion can be a critical factor in mining
sary criteria in mining method selection; the intangible differences method selection stage; the prominence of a factor should not be
between the characteristics of each mine are often recognized by the cause of neglecting other factors. All the factors should be com-
pared considering different mining methods in order to select the
best method. In the last years, Physical and mechanical character-
⇑ Corresponding author. Tel.: +98 22882200. istics of deposit, such as shape, grade and geomechanical features,
E-mail address: [email protected] (H. Karimnia). are used in order to mining method selection. Boshkov and Wright,

http://dx.doi.org/10.1016/j.ijmst.2015.02.010
2095-2686/Ó 2015 Published by Elsevier B.V. on behalf of China University of Mining & Technology.
226 H. Karimnia, H. Bagloo / International Journal of Mining Science and Technology 25 (2015) 225–230

Morison, Laubscher and Hartman suggested some techniques to After the construction of the analytical hierarchy tree, pairwise
mining method selection [1–4]. The suggested techniques are not comparisons are made between criteria in respect of the goal, and
adequate to select the best method secured mine technical issues between the alternatives in respect of the any criterion. Then, pair-
and economic. In this paper, the factors affecting the mining wise comparison matrixes can be provided based on the results
method selection are determined. These factors include shape, derived from the opinion survey. Using these matrixes, the pri-
thickness, depth, slope, RMR and RSS of the orebody, RMR and orities of the alternatives and inconsistency index are calculated.
RSS of the hanging wall and footwall. Then, the priorities of these Finally, alternatives are ranked by their priorities. Further, the deci-
factors are calculated in order to select the best mining method sion maker might express or define a ranking (weighting) for the
for Qapiliq salt mine, Iran, using fuzzy analytical hierarchy process criteria to reflect their importance. There are many forms for
(AHP) technique. AHP is one of the multiple criteria decision mak- expressing the relative importance of criteria, but the most com-
ing techniques that use the viewpoints of experts in order to select mon are: utility preference functions; the analytical hierarchy pro-
the best option from several alternatives. Fuzzy AHP technique is cesses; and fuzzy version of the classical linear weighted average.
used for mining method selection in this mine because of its accu- Notability for any fuzzy decision criteria could be fuzzy or crisp.
rate outputs in the past similar works. This technique selects the The aim of multiple attribute decision making is to derive the best
best mining method using the viewpoints of all available experts alternative as the one that shows the highest degree of satisfaction
considering the criteria and alternatives defined for mining for all preselected attributes and predefined goals. In order to
method selection problem. obtain the best alternative, a ranking process is required. If the rat-
ing for alternative Ak is crisp, there is no problem and the best
alternative is the one with the highest support. When the rating
2. Fuzzy theory
is itself a fuzzy set, a more sophisticated ranking procedure is
required. The focus of this paper is on Yager’s method, which is
Acquiring the information necessary for mining method selec-
general enough to deal with both multiple objectives and multiple
tion is an elaborate process, to say the least, and once obtained,
attribute problems. The Yager method follows the maxim method
data is likely to be ambiguous. In addition, decision makers must
of Bellman and Zadeh, with the improvement of Saaty’s method,
often apply rules of thumb or incorporate their personal intuition
which considers the use of a reciprocal matrix to express the
and judgment when deriving performance measures based on
pair-wise comparison criteria and the resulting eigenvector as sub-
indefinite linguistic concepts, e.g. ‘‘high’’, ‘‘low’’, ‘‘strong’’, ‘‘weak’’,
jective weights [8,9]. The weighting procedure uses exponentials
and ‘‘stable’’. Such terminology is common and is caused by imper-
based on the definition of linguistic hedges, proposed by Zadeh
fectly defined problem attributes. Fuzzy sets have vague bound-
[10]. On describing multiple attribute decision making problems,
aries and are therefore well suited for representing linguistic
only a single objective is considered, namely the selection of the
terms such as ‘‘very’’ or ‘‘somewhat’’ or natural phenomena such
best alternative from a set of alternatives. The decision method
as temperatures. Fuzzy set theory is used to describe fuzzy sets,
assumes the max–min principle approach. Formally, let A = {A1,
and was developed as an alternative to ordinary (crisp) set theory.
A2, . . ., An} be the set of alternatives, C = {C1, C2, . . ., Cm} be the set
Fuzzy logic is used to derive the set membership function for a
of criteria, which can be given as fuzzy sets in the space of alterna-
fuzzy set, which is used for fuzzy logic decision making. The prob-
tives. Hence, the fuzzy set decision is the intersection of all criteria:
lem of constructing meaningful and suitable membership func-
lD(A) = Min{lC1(Ai), lC2(Ai), . . ., lCm (Ai)}. For all (Ai)2A, and the
tions involves a lot of additional research. A number of empirical
optimal decision is yielded by, lD(A⁄) = Max(lD(Ai)), where A⁄ is
ways to establish membership functions for fuzzy sets are known.
the optimal decision. A main difference in this approach is that
the importance of criteria is represented as exponential scalars.
3. Fuzzy analytical hierarchy process This is based on the idea of linguistic hedges. The rationale behind
using weights (or importance levels) as exponents is that the high-
Analytical hierarchy process is one of the most important of the er the importance of criteria, the larger should be the exponent,
multiple criteria decision making techniques that presented by giving the minimum rule. Conversely, the less important a criteri-
Saaty in 1980 [5–7]. This technique investigates complicated prob- on, is the smaller its weight. This seems intuitive. Formally:
lems based on their interactions and resolves them by turning into
  n a1  a2  a o
the simple form. AHP method can be used for solving decision lD Af ¼ Min lC1 ðA1 Þ ; lC 2 ðA1 Þ ; . . . ; lC m ðA1 Þ m ð1Þ
making problems with several alternatives and criteria. This tech-
nique is based on the pairwise comparison between the alterna-
tives and the criteria in respect of the criteria and the goal of the Consider the problem of selecting an alternative from a set of
problem, respectively. For this purpose, analytical hierarchy tree alternatives {A1, A2, A3} for which the set of criteria C1, C2, and C3
is provided as shown in Fig. 1. The first, second and third level of is defined. The judgment scale used is as follows: 1 means equally
the tree diagram are the goal of the problem, the criteria and the important, 3 means weakly more important, 5 means strongly
alternatives, respectively. Considering the type of the problem, more important, 7 means demonstrably more important and 9
the criteria level can be divided into sublevels. means absolutely more important. The values between 2, 4, 6,
and 8 indicate some compromised judgment [7]. Yager [8] suggests
the use of Saaty’s method for pair-wise comparison of the criteria
Goal First level (attributes). A pair-wise comparison of criteria (attributes) could
improve and facilitate the assessment of criteria importance. Saaty
developed a procedure for obtaining a ratio scale for the elements
compared. To obtain the importance, the decision maker is asked
Criterion A Criterion B Criterion C Criterion D Second level to judge the criteria in pair wise comparisons and the values
assigned are Wij = 1/Wij. Having obtained the judgments, an
m  m matrix B is constructed so that bii = 1, and bij = wij and
Alternative A Alternative B Alternative C Third level bji = 1/bij Yager suggests that, with respect to a decision problem,
the use of the resulting eigenvector expresses a decision maker’s
Fig. 1. The analytical hierarchy tree diagram. empirical estimate of the level of importance of alternatives for a
 H. Karimnia, H. Bagloo / International Journal of Mining Science and Technology 25 (2015) 225–230 227

Table 1
Geotechnical properties of rock salt.

Parameter Value
Uniaxial compressive strength (MPa) 18.5
Internal friction angle (°) 55
Cohesion (MPa) 4
Specific gravity (kg/m3) 2100
Elastic modulus (GPa) 5
Poisson’s ratio 0.31

Table 2
Geotechnical properties of rock mass salt.

Parameter Value
Uniaxial compressive strength (MPa) 0.217
Internal friction angle (°) 47.5
Fig. 2. Qapiliq mine location. Cohesion (MPa) 0.22
Specific gravity (kg/m3) 2100
Elastic modulus (GPa) 1.57

Mining method selection

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10

Area mining Stope and pillar Cut and fill Room and pillar

Criteria
Indicator C1 C2 C3 C4 C5 C6 C7 C8 C9 C10
Hanging Hanging
Shape Thickness Ore Footwall Ore Footwall
Criterion Depth Slope wall wall
Ore RMR RSS
Fig. 3. Number of samples taken from a Qapiliq salt mine.

Fig. 4. Analytical hierarchy tree of the mining method selection problem of Qapiliq
salt mine.
given criterion [11]. If C2 and C3 are three and two times as impor-
tant as C1, respectively, and C2 is three times as important as C3, the n o
pair-wise comparison reciprocal matrix will be expressed as lD ðA2 Þ ¼ Min ½0:80:157 ; ½0:950:594 ; ½0:730:249 ¼ 0:92 ð3Þ

n o
lD ðA3 Þ ¼ Min ½0:540:157 ; ½0:320:594 ; ½0:40:249 ¼ 0:51 ð4Þ

Consequently, the optimal solution, corresponding to the max-

imum membership level of 0.92, is given as
 
0:92
lD ðA Þ ¼ ð5Þ
A2

4. Description of the studied site

Hence, the eigenvalues of the reciprocal matrix are
k = [03.0530] and, therefore, kmax = 3.054. The relative weights of Qapiliq salt mine is located near the village of Qapiliq,
the criteria are finally achieved in the eigenvector of the matrix, approximately 45 km west of Khoy city of West Azerbaijan, Iran.
i.e. eigenvector = {0.157, 0.594, 0.249}, with kmax. The eigenvector Fig. 2 illustrates the Location of the Qapiliq mine. Geological struc-
reflects the weights associated with each attribute, feature and ture of mine area consists of salt domes covered by clay and azarit
goal of a decision problem. Thus the exponential weightings are layers whit the thickness of 2–7 m. Salt domes are created during
a1 = 0.157, a2 = 0.594, a3 = 0.249; the final decision expressed in a the salt mass upward running by lateral pressures because of the
membership decision function, can be determined as follows: lD salt lower special gravity than the adjacent rocks and its plasticity.
(A) = Min{lC10.157, lC20.594, lC30.294}. If the relative levels of impor- For early years, salt stone in this mine was extracted using area
tance of criteria for alternative A1 are 0.75, 0.4 and 0.7, respectively, mining method. Because of some economic and technical prob-
those for alternative A2 are 0.8, 0.95 and 0.73, respectively, and lems, another mining method should have been selected to extract
those for alternative A3 are 0.54, 0.32 and 0.4 respectively, then the rest of the deposit. The mining method selection process for
the applicable membership decision functions of alternatives A1, this mine is described in this paper.
A2 and A3, respectively, can be defined as follows: To determine the mechanical properties of intact rock, uniaxial
n o
and three-axis compressive strength testing is done. To test deter-
lD ðA1 Þ ¼ Min ½0:750:157 ; ½0:40:594 ; ½0:70:249 ¼ 0:58 ð2Þ
mining the uniaxial compressive strength of cylindrical samples of
228 H. Karimnia, H. Bagloo / International Journal of Mining Science and Technology 25 (2015) 225–230

Table 3 Table 5
Criteria quantities of Qapiliq salt mine method selection. Criteria priorities.

Criterion Indicator Quantity Criteria Not normal weight Normalized weight

Ore shape C1 Blocky C1 0.90 0.135
Ore thickness (m) C2 60–80 C2 0.87 0.130
Depth (m) C3 10 C3 0.71 0.110
Slope (°) C4 10–25 C4 1.00 0.150
Hanging wall RMR C5 58 C5 0.97 0.145
Ore RMR C6 65 C6 0.73 0.109
Footwall RMR C7 58 C7 0.57 0.085
Hanging wall RSS C8 5.7 C8 0.32 0.048
Ore RSS C9 7.8 C9 0.39 0.058
Footwall RSS C10 6.5 C10 0.21 0.031

Table 4
Scores of phrases applied in questionnaire of the mining method selection problem of
0.16
Qapiliq salt mine.
0.14
Phrase Low Very Medium 0.12
low

Priority
0.10
High Very high 0.08
0.06
Score 1 2 3 4 5 6 7 8 9
0.04
0.02
rock salt in accordance with ASTMD 2938 (Fig. 3) is used in the 0
standard ratio of height to diameter of between 2 and 2.5. C1 C2 C3 C4 C5 C6 C7 C8 C9 C10
To test determining the compressive strength of three axis, Criterion
cylindrical specimens prepared in a special three axis chamber
Fig. 6. Criteria priorities chart.
(Hock-Cell) has been and with identified confining stress, the sam-
ple final resistance of is obtained. Geotechnical properties of the
rock salt in Table 1 obtained. Fig. 4. The goal of this decision making problem is the selection
To estimate geotechnical parameters of rock mass based on the of the best mining method; the criteria include shape, thickness,
results laboratory tests of Hook and Brown failure criterion and depth, slope, RMR and RSS of the orebody, RMR and RSS of the
RocLab software is used. The results in Table 2 are presented. hanging wall and footwall; and the alternatives are the area mining
(A1), stope and pillar (A2), cut and fill (A3), room and pillar (A4)
methods.
5. Mining method selection using fuzzy AHP technique The criteria required to alternatives pairwise comparison are
concluded using some tests and investigations. Table 3 shows the-
For selecting the best mining method for Qapiliq salt mine using se criteria with their indicators.
AHP technique, the analytical hierarchy tree is provided, as shown
in Criteria

Indicator C1 C2 C3 C4 C5 C6 C7 C8 C9 C10
Criterion Shape Thickness Depth Slope Hanging wall Ore Footwall Hanging wall Ore Footwall
Ore RMR RSS

Cl C2 C3 C4 C5 C6 C7 C8 C9 C10

C1 (1,1,1) (1,1.35,2) (2,2.2,3) (0.25,0.4,0.5) (0.25,1.8,4) (1,2,3) (2,3.8,6) (2,3,5) (2,3.5,6) (2,4.4,8)

C2 (0.5,0.9,2) (1,1,1) (2,2.9,5) (0.33,0.63,1) (0.33,1.2,2) (1,2,3) (1,3,4) (1,2.3,4) (1,2,4) (2,6,9)

C3 (0.33,0.47,0.5) (0.2,0.39,0.5) (1,1,1) (0.17,0.27,0.5) (0.17,0,52,2) (0.33,1.2,2) (0.5,1.6,3) (1,2.2,4) (1,2.2,5) (1,3.8,8)

C4 (2,2.7,4) (1,1.8,3) (2,3.9,6) (1,1,1) (0.5,2.15,4) (2,3,4) (2,4.7,8) (2,3.5,5) (2,3.5,4) (2,4.6,6)

C5 (0.25,1.43,4) (0.5,1.5,3) (0.5,2.9,6) (0.5,0.75,2) (1,1,1) (0.5,2.5,6) (0.5,4.9,8) (2,3.3,4) (2,4.6,8) (2,5.5,8)

C6 (0.5,0.53,1) (0.33,0.57,1) (0.5,1.2,3) (0.25,0.37,0.5) (0.16,1,2) (1,1,1) (2,2,2) (0.5,1.9,4) (1,2.6,3) (1,4.4,8)

C7 (0.17,0.29,0.5) (0.25,0.4,1) (0.33,0.97,2) (0.12,0.26,0.5) (0.12,0.48,2) (0.5,0.5,0.5) (1,1,1) (1,2,4) (1,1.8,2) (2,3.2,5)

C8 (0.2,0.38,0.5) (0.5,0.52,1) (0.25,0.62,1) (0.17,0.34,0.5) (0.25,0.32,0.5) (0.25,0.82,1) (0.33,0.67,1) (1,1,1) (0.33,1.7,1) (0.5,1.8,4)

C9 (0.17,0.31,0.5) (0.25,0.57,1) (0.2,0.76,1) (0.25,0.3,0.5) (0.12,0.29,0.5) (0.33,0.43,1) (0.5,0,6,1) (0.5,1.7,3) (1,1,1) (1,2,3)

C10 (0.13,0.29,0.5) (0.11,0.33,0.5) (0.12,0.38,1) (0.17,0.25,0.5) (0.12,0.27,0.5) (0.12,0.34,1) (0.2,0.34,0.5) (0.25,1.2,2) (0.5,0.6,1) (1,1,1)

Fig. 5. Criteria pairwise comparison matrix.

 H. Karimnia, H. Bagloo / International Journal of Mining Science and Technology 25 (2015) 225–230 229

m
Cl Al A2 A3 A4 ∑M
j =1
j
gi Alternative
Not normal
weight
Normalized
weight
Al (1,1,1) (0.12,0.39,1) (0.25,0.75,2) (0.33,1.8,4) (1.7,3.94,8) Al 1.00 0.27
A2 (1,3.5,8) (1,1,1) (0.33,1.3,3) (0.5,1.6,4) (2.83,7.4,16) A2 1.00 0.27
A3 (0.5,2.1,4) (0.33,1.7,3) (1,1,1) (0.33,1.6,4) (2.16,6.4,12) A3 0.84 0.22
A4 (0.33,1.1,3) (0.25,0.82,2) (0.25,1.3,3) (1,1,1) (1.83,4.22,9) A4 0.89 0.24

m
Cl Al A2 A3 A4 ∑M
j =1
j
gi Alternative
Not normal
weight
Normalized
weight
Al (1,1,1) (1,3.1,6) (1,2.9,7) (2,3.1,5) (5,10.1,19) Al 1.00 0.32
A2 (0.17,0.41,1) (1,1,1) (2,3.1,5) (1,2.7,4) (4.17,7.21,11) A2 0.89 0.28
A3 (0.25,0.46,1) (0.2,0.35,0.5) (1,1,1) (0.5,1.4,3) (1.95,3.21,5.5) A3 0.62 0.19
A4 (0.2,0.37,0.5) (0.25,0.48,1) (0.33,1.3,3) (1,1,1) (1.78,3.15,5.5) A4 0.62 0.19

m
Cl Al A2 A3 A4 ∑M
j =1
j
gi Alternative
Not normal
weight
Normalized
weight
Al (1,1,1) (0.12,0.2,0.33) (0.25,0.32,0.5) (0.14,0.28,0.5) (1.51,1.8,2.33) Al 0.18 0.06
A2 (3,5.5,8) (1,1,1) (0.33,1.8,3) (1,2.8,6) (5.33,11.1,18) A2 1.00 0.34
A3 (2,3.5,6) (0.33,2.2,3) (1,1,1) (0.33,1.2,2) (3.66,7.9,12) A3 0.88 0.30
A4 (2,4.4,8) (0.17,0.53,1) (0.5,1.35,3) (1,1,1) (3.67,7.28,13) A4 0.88 0.30

m
Cl Al A2 A3 A4 ∑M
j =1
j
gi Alternative
Not normal
weight
Normalized
weight
Al (1,1,1) (0.17,0.3,0.5) (1,2.2,3) (0.25,0.49,1) (2.24,3.99,5.5) Al 0.40 0.16
A2 (2,3.7,6) (1,1,1) (3,3.8,6) (1,1.7,3) (7,10.2,16) A2 1.00 0.41
A3 (0.33,0.5,1) (0.17,0.28,0.33) (1,1,1) (0.33,0.7,3) (1.83,2.48,5.33) A3 0.34 0.14
A4 (1,2.3,4) (0.33,0.68,1) (0.33,2.1,3) (1,1,1) (2.66,6.08,9) A4 0.70 0.29

m
Cl Al A2 A3 A4 ∑M
j =1
j
gi Alternative
Not normal
weight
Normalized
weight
Al (1,1,1) (0.17,1.2,2) (0.25,2.7,6) (0.33,1.9,4) (1.75,6.8,13) Al 0.96 0.27
A2 (0.5,1.8,6) (1,1,1) (2,3.2,4) (1,1.7,3) (4.5,7.7,14) A2 1.00 0.28
A3 (0.17,1.1,4) (0.25,0.34,0.5) (1,1,1) (0.25,0.42,1) (1.67,2.86,6.5) A3 0.71 0.20
A4 (0.25,0.87,3) (0.33,0.68,1) (1,2.7,4) (1,1,1) (2.58,5.25,9) A4 0.88 0.25

m
Cl Al A2 A3 A4 ∑M
j =1
j
gi Alternative
Not normal
weight
Normalized
weight
Al (1,1,1) (0.17,0.3,1) (0.25,0.85,6) (0.25,0.32,0.5) (1.67,2.47,8.5) Al 0.63 0.21
A2 (2,3.7,6) (1,1,1) (2,3.4,5) (1,1.2,2) (6,9.3,14) A2 1.00 0.33
A3 (0.17,2.5,4) (0.2,0.32,0.5) (1,1,1) (0.2,0.39,1) (1.57,4.21,6.5) A3 0.48 0.16
A4 (2,3.3,4) (0.5,0.9,1) (2,2.8,5) (1,1,1) (5.5,8,11) A4 0.93 0.31

m
Cl Al A2 A3 A4 ∑M
j =1
j
gi Alternative
Not normal
weight
Normalized
weight
Al (1,1,1) (0.25,0.33,0.5) (0.25,0.42,1) (0.17,0.37,0.5) (1.67,2.12,3) Al 0.09 0.03
A2 (2,3.3,4) (1,1,1) (2,3.1,4) (1,1.5,4) (6,8.9,13) A2 1.00 0.37
A3 (1,2.9,4) (0.25,0.33,0.5) (1,1,1) (0.25,1.2,2) (2.5,5.43,7.5) A3 0.72 0.27
A4 (2,3.1,6) (0.25,0.82,1) (0.5,1.6,4) (1,1,1) (3.75,6.52,12) A4 0.87 0.32

m
Cl Al A2 A3 A4 ∑M
j =1
j
gi Alternative
Not normal
weight
Normalized
weight
Al (1,1,1) (0.17,0.25,0.33) (0.25,0.35,0.5) (0.17,0.32,0.5) (1.59,1.92,2.33) Al 0.14 0.05
A2 (3,4.2,6) (1,1,1) (0.33,2.5,4) (1,1.6,5) (5.33,9.3,16) A2 1.00 0.35
A3 (2,3,4) (0.25,0.77,3) (1,1,1) (0.33,1.8,3) (3.58,6.57,11) A3 0.85 0.30
A4 (2,3.7,6) (0.2,0.82,1) (0.33,0.83,3) (1,1,1) (3.53,6.35,11) A4 0.85 0.30

m
Cl Al A2 A3 A4 ∑M
j =1
j
gi Alternative
Not normal
weight
Normalized
weight
Al (1,1,1) (0.14,0.27,0.33) (0.25,0.4,0.5) (0.14,0.27,0.5) (1.53,1.94,2.33) Al 0.08 0.03
A2 (3,4.2,7) (1,1,1) (0.5,3.25,4) (1,2.2,4) (5.5,10.65,16) A2 1.00 0.39
A3 (2,2.7,4) (0.25,0.47,2) (1,1,1) (0.25,1.3,2) (3.5,5.47,9) A3 0.68 0.26
A4 (3,4.2,7) (0.25,0.54,1) (0.5,1.4,4) (1,1,1) (4.75,7.14,13) A4 0.82 0.32

m
Cl Al A2 A3 A4 ∑M
j =1
j
gi Alternative
Not normal
weight
Normalized
weight
Al (1,1,1) (0.14,0.24,0.5) (0.25,0.36,0.5) (0.14,0.33,0.5) (1.53,1.93,2.5) Al 0.12 0.05
A2 (2,4.8,7) (1,1,1) (2,3.5,6) (1,2.5,5) (6,11.8,19) A2 1.00 0.40
A3 (2,2.9,4) (0.17,0.32,0.5) (1,1,1) (0.25,0.44,1) (3.42,4.66,6.5) A3 0.54 0.22
A4 (2,3.6,7) (0.2,0.54,1) (1,2.4,4) (1,1,1) (4.2,7.54,13) A4 0.82 0.33

Fig. 7. Alternatives pairwise comparison matrixes in respect of criteria.

230 H. Karimnia, H. Bagloo / International Journal of Mining Science and Technology 25 (2015) 225–230

Table 6
Final score alternatives in terms of criteria.

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 Overall
Weight 0.135 0.130 0.110 0.150 0.145 0.109 0.085 0.048 0.058 0.031 Priority
A1 0.27 0.32 0.06 0.16 0.27 0.21 0.03 0.05 0.03 0.05 0.179
A2 0.27 0.28 0.34 0.41 0.28 0.33 0.37 0.35 0.39 0.40 0.298
A3 0.22 0.19 0.30 0.14 0.20 0.16 0.27 0.30 0.26 0.22 0.214
A4 0.24 0.19 0.30 0.29 0.25 0.31 0.32 0.30 0.32 0.33 0.274

this mine are room and pillar, area mining and cut and fill methods,
0.30
respectively.
0.25

Room and pillar

Stope and pillar
Priority

0.20
6. Conclusions
Cut and fill

0.15
Are mining

0.10 Mining method selection is one of the most important and the
0.05 most essential of decisions of a mining project that have a sig-
0 nificant influence on the all of the mine decision making problems.
A1 A2 A3 A4
Suitable mining method selection for a deposit requires consider-
Alternative ing a series of various criteria. Several selection methods such as
Fig. 8. Overall priorities chart of the alternatives.
Nikolas, modified Nikolas and UBC have been applied for many
years to investigate mining methods. These methods have not con-
sidered the priority factor of the criteria used for mining method
Considering the analytical hierarchy tree provided for Qapiliq selection. In this paper, the best mining method for Qapiliq salt
salt mine, a questionnaire was prepared and was given to the asso- mine has been selected using fuzzy analytical hierarchy process
ciated experts. The scores of phrases applied in questionnaire are technique based on the viewpoints of the experts considering ten
shown in Table 4. criteria and four alternatives. After calculating the priority of the
Fig. 5 indicates the 10  10 matrix of paired comparisons of cri- alternatives, the feasible mining methods for this mine have been
teria based on the decision makers empirical estimation. According ranked. The results show that the stope and pillar method with the
to the decision committee, it has several decisions, a couple of priority of 0.298 is the best for Qapiliq salt mine exploitation.
entries is a triangular fuzzy number in this matrix of the first com-
ponent minimum Polls, its second component is average Polls and References
the third component is the maximum Polls. Criteria priorities are
[1] Boshkov SH, Wright FD. Basic and parametric criteria in the selection, design
calculated using eigenvector method.
and development of underground mining systems. SME mining engineering
As shown in Table 5 and Fig. 6, slope is the most important cri- handbook. New York: SME-AIME; 1973.
terion for Qapiliq salt mine method selection based on the opinion [2] Hartman HL. Introductory mining engineering. New York: Wiley; 1987.
[3] Laubscher DH. Selection of mass underground mining methods. Design and
survey. Finally, with calculated final weight vector, not normal
operation of caving and sublevel stopping mines. New York: SME-AIME; 1981.
weight and normalized each criterion in Table 5 is presented. [4] Morrison RGK. A philosophy of ground control. Montreal: McGill University;
Alternatives pairwise comparison matrixes in respect of criteria 1976.
are also calculated based on the geometric mean of the results of [5] Saaty TL. The analytic hierarchy process. New York: McGraw-Hill; 1980.
[6] Saaty TL. Axiomatic foundation of the analytic hierarchy process. Manage Sci
the expert’s viewpoint. These matrixes are shown in Fig. 7 A1, A2, 1986;32(7):841–55.
A3 and A4 in these matrixes are area mining, stope and pillar, cut [7] Saaty TL, Vargas LG. Decision making in economic, political, social, and
and fill and room and pillar methods, respectively. After the calcu- technological environments with the analytic hierarchy process. US: RWS
Publications; 1994.
lation and determination of final weight alternatives, the summa- [8] Yager RR. Fuzzy decision making including unequal objectives. Fuzzy Sets Syst
rized calculated results of importance coefficients in the matrix of 1978;1:87–95.
paired comparisons and Points for each of the alternatives [9] Bellman RE, Zadeh LA. Decision making in a fuzzy environment. Manage Sci
1970;17(4):141–64.
described in [10] Zadeh LA. Outline of a new approach to the analysis of complex systems and
Table 6 is calculated. The final score for each of the alternatives decision process. IEEE Trans Syst 1973;3(1):28–44.
is calculated by multiplying the weight of each alternative in its [11] Basetin A, Kesimal A. The study of a fuzzy set theory for the selection of an
optimum coal transportation system from pit to the power plant. Int J Surf
criteria. Finally, the considering the alternatives priorities, stope
Min, Reclamation Environ 1999;13:97–101.
and pillar with the overall priority of 0.298, is the best mining
method for Qapiliq salt mine (Fig. 8). Other mining methods for