Professional Documents
Culture Documents
a r t i c l e i n f o a b s t r a c t
Article history: Underground mining method selection is one of the most important decisions that mining engineers have
Received 28 January 2008 to make. Choosing a suitable underground mining method to extract a mineral deposit is very important
Received in revised form 3 July 2008 in terms of economics, safety and productivity of mining operations. In real life, underground mining
Accepted 29 July 2008
method selection is one of the multiple attribute decision making (MADM) problems and decision makers
Available online 17 September 2008
have always some difficulties in making the right decision in the multiple criteria environment. Analytic
hierarchy process (AHP) and Yager’s methods are the MADM tools and can be used for selection of the
Keywords:
best underground mining method by considering the problem criteria. In this study, a computer program
Analytic hierarchy process
Multiple attribute decision making
(UMMS) based on the AHP and the Yager’s method was developed to analyze the underground mining
Yager’s method method selection problems and produce the best underground mining method swiftly for different
Underground mining method selection deposit shapes and ore bodies.
Ó 2008 Elsevier Ltd. All rights reserved.
0886-7798/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tust.2008.07.003
174 S. Alpay, M. Yavuz / Tunnelling and Underground Space Technology 24 (2009) 173–184
method selection problem by using the Yager’s method and they Table 1
used Satty’s analytic hierarchy process (AHP) method for pair-wise Scale for pair-wise comparisons (Saaty, 1980)
comparison of the criteria. Elevli et al. (2002) selected a new verti- Relative Definition Explanation
cal shaft or ramp system by comparing the weighted alternative intensity
criteria for a small-scale underground mine on the basis of total 1 Of equal value Two requirements are of equal value
investment cost, ore transport unit cost and net present value of 3 Slightly more value Experience slightly favors one
overall project for various depths. Kesimal and Bascetin (2002) requirement over another
5 Essential or strong value Experience strongly favors one
used the Yager’s method for solving equipment selection problem requirement over another
in open pit mine. Samanta et al. (2002) used the AHP method for 7 Very strong value A requirement is strongly favored and its
selection of open cast mining equipment. Bitarafan and Ataei dominance is demonstrated in practice
(2004) solved the similar problem by using the Yager’s method 9 Extreme value The evidence favoring one over another is
of the highest possible order of
with Satty’s AHP method and they also used fuzzy dominance
affirmation
method in their analysis. Elevli and Demirci (2004) selected most 2, 4, 6, 8 Intermediate values When compromise is needed
suitable underground ore transport system for a chromate mine between two adjacent
by using the one of the multiple attribute decision making (MADM) judgments
method namely Preference Ranking Organization MeTHod for
Enrichment Evaluation (PROMETHEE). Kazakidis et al. (2004) used
the AHP method and analyzed five different mining scenarios such Since the comparison is based on the subjective evaluation, a
as drilling technology investment analysis, ground support design, consistency ratio is required to ensure the selection accuracy.
tunneling systems design, shaft location selection and mine plan- The consistency index (CI) of the comparison matrix is computed
ning risk assessment. Ataei (2005) used the AHP method for the as follows:
problem of selection of a new alumina cement plant location in
CI ¼ ðkmax nÞ=ðn 1Þ ð2Þ
East-Azerbaijan province of Iran. Acaroglu et al. (2006a, 2006b)
used the Yager’s and the AHP method for selection of roadheaders where kmax is maximal or principal eigenvalue and n is the matrix
in tunneling applications. Bascetin et al. (2006) developed a com- size. The consistency ratio (CR) is calculated as
puter program using the Yager’s method for equipment and mining CR ¼ CI=RI ð3Þ
method selection in mining. Yavuz et al. (2008) used the AHP
method for selection of optimum support type in the main haulage where ‘‘RI” denotes random consistency index. Random consistency
road in Lignite colliery. indices are given in Table 2 (Saaty, 2000).
In this paper, a computer program (UMMS) based on the AHP As a general rule, a consistency ratio of 0.10 or less is considered
and the Yager’s method for underground mining method selection acceptable. In practice, however, consistency ratios exceeding 0.10
was developed. The UMMS provides two different AHP models so occur frequently.
that the decision makers can analyze the selection problem based
on either only main criteria or main criteria with their sub-criteria. 3.2. FMADM model
The UMMS also provides a Yager’s method which is one of the fuz-
zy multiple attribute decision making (FMADM) methods as an FMADM methods have been developed due to the lack of preci-
alternative way to analyze the problem. Another important func- sion in assessing the relative importance of attributes and the per-
tion provided by the UMMS is the sensitivity analysis. The decision formance ratings of alternatives with respect to an attribute. The
makers can carry out the sensitivity analysis whenever he/she problem of FMADM is to select/prioritize/rank a finite number of
needs to observe the sensitivity of final solution against the varia- alternatives by evaluating a group of predetermined criteria. Thus,
tions in the main criteria considered. to solve this problem, an evaluation method to rate and rank, in or-
der of preference, the set of alternatives must be constructed (Chen
and Klein, 1997).
3. Theory review
Although a large number of FMADM methods have been ad-
dressed in the literature, the focus of this paper is on Yager’s
3.1. AHP model
(1978) method. This method is general enough to deal with both
multiple objectives and multiple attribute problems and follows
The AHP method developed by Saaty (1980) gives an opportu-
the max–min method of Bellman and Zadeh (1970), with the
nity to represent the interaction of multiple factors in complex
improvement of Saaty’s method, which considers the use of a reci-
unstructured situations. The method is based on the pair-wise
procal matrix to express the pair-wise comparison of the criteria
comparison of components with respect to attributes and alterna-
and the resulting eigenvector as subjective weights. The weighting
tives. A pair-wise comparison matrix n n is constructed, where n
procedure uses exponentials based on the definition of linguistic
is the number of elements to be compared. The method is applied
hedges, proposed by Zadeh (1973).
for the hierarchy problem structuring (Saaty, 2000).
On describing multiple attribute decision making problems,
After the hierarchy structuring the pair-wise comparison matrix
only a single objective is considered, namely the selection of the
is constructed for each level, where a nominal discrete scale from 1
best alternative from a set of alternatives. The decision method as-
to 9 (Table 1) is used for the evaluation (Saaty, 1980).
sumes the max–min principle approach. Formally, let
The next step is to find the relative priorities of criteria or alter-
A = {A1, A2, . . . , Am} be the set of alternatives, C = {C1, C2, . . . , Cn} be
natives implied by this comparison. The relative priorities are
the set of criteria, which can be given as fuzzy sets in the space
worked out using the theory of eigenvector. For example, if the pair
of alternatives. Hence, the fuzzy set decision is the intersection of
comparison matrix is A, then,
all criteria: lD ðAÞ ¼ min½lc1 ðAÞ; lC 2 ðAÞ; . . . ; lC n ðAÞ. For all (Ai) e A,
ðA kmax IÞ w ¼ 0 ð1Þ and the optimal decision is yielded by, lD ðA Þ ¼ maxA lD ðAÞ, where
A* is the optimal decision. Main difference in this approach is that
To calculate the eigenvalue ‘‘kmax” and eigenvector the importance of criteria is represented as exponential scalars.
w = (w1, w2, . . . , wn), weights can be estimated as relative priorities The rationale behind using weights (or importance levels) as expo-
of criteria or alternatives (Saaty, 2000). nents is that the higher the importance of criteria, the larger should
S. Alpay, M. Yavuz / Tunnelling and Underground Space Technology 24 (2009) 173–184 175
Table 2
Consistency indices of randomly generated reciprocal matrices (Saaty, 2000)
Order of matrix 1, 2 3 4 5 6 7 8 9 10 11 12 13 14 15
RI value 0.00 0.58 0.90 1.12 1.24 1.32 1.41 1.45 1.49 1.51 1.48 1.56 1.57 1.59
be the exponent, giving the minimum rule (Riberio, 1996). quite and this also makes the time needed to reach a solution much
Formally: longer. If the time is restricted and/or the six main group criteria
n o without their sub-criteria are enough to make a decision, the sec-
lD ðAÞ ¼ min ½lC1 ðAÞa1 ; ½lC2 ðAÞa2 ; . . . ; ½lCn ðAÞan ð4Þ ond AHP model can be used. The first level of the both models is
the general object. In the first model, six main group criteria are
Yager suggests the use of Saaty’s method for pair-wise comparison
placed in the second level and their sub-criteria are placed in the
of the attributes and the use of the resulting eigenvector expresses a
third level. In the second model, only main group criteria are con-
decision maker’s empirical estimate of the level of importance of
sidered in the decision making process and placed in the second le-
alternatives for a given criterion.
vel of the model. The last level of either of the models is the
alternatives.
4. Methodology
(c) Determination of the set of alternatives: The alternatives of the
The approach proposed in this study comprises the following AHP models and FMADM are determined based on the UBC
steps: approach by Miller-Tait et al. (1995). The UBC approach is
simply a modified version of the Nicholas approach and
(a) Determination of the method: The method may be AHP or numerically ranks deposit characteristics of ore geometry
FMADM. and rock mechanics characteristics of the ore zone, footwall
(b) Determination of the AHP model type (AHP method only): The and hanging wall. The rankings are then summed together
decision maker should select one of two AHP models. The with the higher rankings being the more favorable or likely
hierarchy structures of the AHP models are shown in Figs. methods. Each ranking consists of a number ‘‘0 to 6” or
1 and 2., respectively. ‘‘10, 49”. ‘‘49” completely eliminates a mining method
from being feasible. A value of ‘‘0” strongly suggests that this
Total of 36 criteria which have been classified in six main characteristic makes that particular mining technique less
groups by Hartman and Mutmansky (2002) are proposed by the attractive. The value ‘‘6” indicates a very favorable charac-
first model to analyze an UMMS problem in details. These criteria teristic for that particular mining method. To determine
are given in Table 3. On the other hand, when the first model is se- the set of valid alternatives, the UMMS uses UBC final
lected, the number of pair-wise comparison matrices increases
Fig. 1. First AHP model including all criteria and their sub-criteria.
Table 4
Ranking of geometry/grade distribution for different mining methods (Miller-Tait et al.,1995)
Mining method General shape Ore thickness Ore plunge Grade distribution Depth
M T/P I VN N I T VT F I S U G E SH I D
Open pit mining 4 2 3 1 2 3 4 4 3 3 1 3 3 2 4 0 49
Block caving 4 2 0 49 49 0 3 4 3 2 4 3 2 2 2 3 3
Sublevel stoping 3 4 1 10 1 3 4 3 2 1 4 4 4 3 3 4 2
Sublevel caving 3 4 1 49 49 0 4 4 1 1 4 3 2 2 3 2 2
Longwall mining 49 4 49 4 3 0 49 49 4 0 49 4 1 0 2 2 3
Room and pillar 0 4 2 4 3 1 49 49 4 0 49 4 2 0 3 3 2
Shrinkage stoping 0 4 2 4 4 0 49 49 49 0 4 3 2 2 3 3 2
Cut and fill stoping 1 4 4 3 4 4 1 0 1 3 4 2 3 4 2 3 4
Top slicing 1 2 0 1 1 0 2 1 4 2 0 2 1 1 2 1 1
Square set stoping 0 1 4 4 3 2 0 0 2 3 2 0 1 3 1 1 2
M = Massive VN = Very narrow F = Flat U = Uniform S = Shallow
T/P = Tabular or platy N = Narrow I = Intermediate G = Gradational I = Intermediate
I = Irregular I = Intermediate S = Steep E = Erratic D = Deep
T = Thick
VT = Very thick
S. Alpay, M. Yavuz / Tunnelling and Underground Space Technology 24 (2009) 173–184 177
Table 5
Rock mechanics characteristics-rock mass rating (Miller-Tait et al.,1995)
RMR ratings: very weak (VW) = 0–20, weak (W) = 21–40, moderate (M) = 41–60, strong (S) = 61–80, very strong (VS) = 81–100.
5.3.2. Altersel
This routine provides a user interface with which the decision
Table 1. The number of comparisons is n (n 1) 2, maker can perform the UBC approach to generate the valid alterna-
where n is the number of elements with the considerations tives and return the appropriate set of them to the UMMS. This
that diagonal elements are 1. The other comparison values user interface can also be used for getting the detailed information
are simply the reciprocals of the earlier comparisons. An about the alternatives provided by GETINFO subroutine.
example of data file with five elements is given in Table 7.
(e) Change value in percent: This value denotes a change 5.3.3. Getinfo
(increase or decrease) in the weights of the main criteria This routine accesses UMMSDAT and provides the decision
and used for the sensitivity analysis. makers the information stored in the tables by a user interface.
UMMS
METODSEL MODSEL
SAVEMTRX
DAT
GETINFO
COMPARA PERFORSENSITIV
MULTMTRX
FINSOL
Spatial characteristics main Size Attitude Depth Regularity Existence 5.3.9. Procfuzzy
criterion This routine asks decision makers for the pair-wise comparisons
Size 1 1/5 1/7 1/3 1/4 of the main criteria and membership levels of each criterion and
Attitude 5 1 1/5 3 2 performs Yager’s FMADM method by calling FINSOL routine to pro-
Depth 7 5 1 7 4
Regularity 3 1/3 1/7 1 1/2
duce the final ratings of the alternatives. The ratings are then re-
Existence 4 1/2 1/4 2 1 turned to the UMMS.
5.3.10. Procahp
This routine asks decision makers for constructing the pair-wise
The routine can also manipulate the information by adding, delet- comparison matrices of the alternatives and criteria related to the
ing and updating operations by the decision makers. selected AHP model. By those matrices, the routine performs the
AHP method and calls the FINSOL routine to produce the final rat-
5.3.4. Loadmtrx ings of the alternatives. The ratings are then returned to the UMMS.
This routine accesses any data file which was already saved to
disk by SAVEMTRX routine and loads the data in the data file into 5.3.11. Finsol
the corresponding comparison matrices in the AHP method or into This routine computes the final rating of the alternatives and
the corresponding comparison matrix at the first level of the ranks them as well. For the FMADM method, the ratings are com-
FMADM method. puted by using the max–min Bellman and Zadeh (1970) principle.
Computed ratings in a vector form are returned to corresponding
caller routine: PROCAHP or PROCFUZZY.
5.3.5. Savemtrx The UMMS takes the final rating and proposes the decision ma-
This routine can save any comparison matrix data as a data file ker the alternative with the highest rating as the most preferred
into the disk. underground mining method. After that time, the decision maker
can either accept the method proposed by the UMMS or make a
5.3.6. Multmtrx sensitivity analysis on the final solution.
Multiplying of matrices is the most used operation in the AHP
method. On the other hand, this operation is also used for the
FMADM method. This routine takes two matrices as parameters 6. Case study: selection of underground mining method for
and returns the multiplied matrix. Eskisehir-Karaburun chromite ore
ing, and top slicing. Although the open pit mining method is
included in the available alternatives, the decision maker will be
asked for whether he/she let this alternative still remain in the va-
lid alternatives set or exclude it from the set. If the decision maker
decides to exclude it, the open pit mining method will no longer be
available further in the decision process and remaining methods
will stay in the alternatives set. For the problem, the open pit min-
ing method was excluded and the AHP method was performed.
The current alternatives of the model are A1: shrinkage stoping,
A2: cut and fill stoping, A3: sublevel stoping, A4: square set stoping
and A5: top slicing. The pair-wise comparisons of the main criteria
were made and given in Fig. 6. It is apparent that spatial character-
istics of the deposit is the most important criterion (priority of
0.429), then economic considerations follows (priority of 0.265).
On the other hand, the pair-wise comparison of the alternatives
based on each criterion should also be performed. Thus, six matri-
ces were formed. As there are five alternatives, the matrix order
was 5 5. The pair-wise comparison of the alternatives by each
criterion is presented in Tables 9–14. The values in the tables were
captured from the corresponding pair-wise comparison matrices of
the UMMS. It is readily observed from the tables that the most suit-
able alternatives are A4, A4, A4, A5, A2 and A5 when judged by the
Fig. 4. The initial windows of the UMMS.
criteria of C1, C2, C3, C4, C5 and C6, respectively.
The overall rating of each alternative is calculated by summing
the product of the relative priority of each criterion and the relative
model, we also determine the criteria to be used for the decision priority of the alternative considering the corresponding criteria. For
making process. example, the overall rating of alternative A1 = (0.429
To determine the valid alternatives for the problem, appropriate 0.095) + (0.046 0.097) + (0.082 0.085) + (0.265 0.125) + (0.149
selections according to the problem parameters given in Table 8 0.115) + (0.028 0.157) = 0.107. The overall ratings calculated by
are made in the UBC rating window of the UMMS as shown in the UMMS are presented in Fig. 7. From Fig. 7, it is obvious that A4
Fig. 5. At this point, it should be noted that the rock mass rating (square-set stoping) with a rating of 0.295 is the most preferred min-
(RMR) values were updated according to Bieniawski (1989). It ing method. In Table 15, all calculated priorities and final ratings are
should also be noted that the rock substance strength (RSS) values summarized.
were calculated by dividing each uniaxial strength of ore, hanging As the comparisons are based on the subjective evaluation, the
wall and footwall to principal stress and principal stress values CR values should be calculated using Eq. (3) to ensure the selection
were calculated by multiplying each gravity for ore, hanging wall accuracy. Table 16 presents kmax, CI, RI, and CR of the correspond-
and footwall with 100 m average depth. ing matrices. For all matrices, the CI and CR are less than 0.10. So,
From Fig. 5, available alternatives are open pit mining, shrink- the logically substantiated decision of constructing the pair-wise
age stoping, cut and fill stoping, sublevel stoping, square set stop- comparison of the criteria or alternatives was made.
Table 9 Table 12
Pair-wise comparison matrix of alternatives for spatial characteristics of deposit (C1) Pair-wise comparison matrix of economic considerations (C4)
A1 A2 A3 A4 A5 Priorities A1 A2 A3 A4 A5 Priorities
A1 1 1/3 3 1/5 1/2 0.095 A1 1 1/3 3 2 1/4 0.125
A2 3 1 4 1/4 2 0.215 A2 3 1 6 5 1/3 0.284
A3 1/3 1/4 1 1/6 1/3 0.051 A3 1/3 1/6 1 1/2 1/5 0.051
A4 5 4 6 1 3 0.488 A4 1/2 1/5 2 1 1/6 0.073
A5 2 1/2 3 1/3 1 0.150 A5 4 3 5 6 1 0.467
kmax = 5.208, CI = 0.052 and CR = 0.046 6 0.1. kmax = 5.217, CI = 0.054 and CR = 0.048 6 0.1.
Table 10 Table 13
Pair-wise comparison matrix of alternatives for geologic and hydrologic conditions Pair-wise comparison matrix of technological factors (C5)
(C2)
A1 A2 A3 A4 A5 Priorities
A1 A2 A3 A4 A5 Priorities A1 1 1/5 3 2 1/4 0.115
A1 1 1/2 2 1/4 1/3 0.097 A2 5 1 6 4 3 0.476
A2 2 1 3 1/3 1/2 0.160 A3 1/3 1/6 1 1/2 1/5 0.052
A3 1/2 1/3 1 1/5 1/4 0.062 A4 1/2 1/4 2 1 1/3 0.089
A4 4 3 5 1 2 0.417 A5 4 1/3 5 3 1 0.267
A5 3 2 4 1/2 1 0.263
kmax = 5.239, CI = 0.060 and CR = 0.053 6 0.1.
kmax = 5.068, CI = 0.017 and CR = 0.015 6 0.1.
Table 11 Table 14
Pair-wise comparison matrix of geotechnical properties (C3) Pair-wise comparison matrix of environmental concerns (C6)
A1 A2 A3 A4 A5 Priorities A1 A2 A3 A4 A5 Priorities
A1 1 1/5 2 1/3 1/2 0.085 A1 1 1/3 3 4 1/4 0.157
A2 5 1 7 1/2 6 0.372 A2 3 1 5 2 1/2 0.269
A3 1/2 1/7 1 1/5 1/3 0.050 A3 1/3 1/5 1 1/2 1/6 0.055
A4 3 2 5 1 4 0.382 A4 1/4 1/2 2 1 1/3 0.095
A5 2 1/6 3 1/4 1 0.111 A5 4 2 6 3 1 0.423
kmax = 5.301, CI = 0.075 and CR = 0.067 6 0.1. kmax = 5.348, CI = 0.087 and CR = 0.078 6 0.1.
6.2. Sensitivity analysis priority of one of the criteria increases, the priorities of the remain-
ing criteria must decrease proportionately, and the global priorities
A sensitivity analysis can be performed to see how sensitive the of the alternatives must be recalculated. Sensitivity analysis can
alternatives will change with the importance of the criteria. As the also be used to determine the most important or critical criterion
S. Alpay, M. Yavuz / Tunnelling and Underground Space Technology 24 (2009) 173–184 181
Table 15 spatial characteristics of the deposit was increased by 8%. This case
Overall result/final matrix was captured from the corresponding sensitivity window of the
A1 A2 A3 A4 A5 Priority UMMS and illustrated in Fig. 8.
It should also be noted that cut and fill stoping method was also
C1 0.095 0.215 0.051 0.488 0.150 0.429
C2 0.097 0.160 0.062 0.417 0.263 0.046 the best alternative when the decreasing of 19% in the priority of
C3 0.085 0.372 0.050 0.382 0.111 0.082 technological factors and the decreasing of 21% in the priority of
C4 0.125 0.284 0.051 0.073 0.467 0.265 economic considerations occur. It can be concluded from the sen-
C5 0.115 0.476 0.052 0.089 0.267 0.149 sitivity analysis that the final solution of the proposed AHP model
C6 0.157 0.269 0.055 0.095 0.423 0.028
is mainly sensible to increasing of spatial characteristics of the de-
Overall 0.107 0.284 0.052 0.295 0.261
posit and decreasing of technological factors and economic
considerations.
More or More or
Very Low Low Medium Less High High Very High
Less Low
(VL) (L) (M) (MLH) (H) (VH)
(MLL)
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Fig. 9. Linguistic model for fuzzy numbers (Chen and Klein, 1997).
Table 17 Table 18
Membership level of each criterion Membership decision function of each criterion by Yager’s method
A1 A2 A3 A4 A5 A1 A2 A3 A4 A5
C1 0.35 0.5 0.2 0.95 0.5 C1 0.637 0.743 0.501 0.978 0.743
C2 0.2 0.35 0.05 0.95 0.65 C2 0.929 0.953 0.871 0.998 0.980
C3 0.2 0.65 0.05 0.8 0.35 C3 0.876 0.965 0.782 0.982 0.918
C4 0.2 0.5 0.2 0.35 0.95 C4 0.653 0.832 0.653 0.757 0.986
C5 0.35 0.65 0.05 0.5 0.5 C5 0.855 0.938 0.640 0.902 0.902
C6 0.2 0.5 0.2 0.35 0.8 C6 0.956 0.981 0.956 0.971 0.994
yielding the result: lD(A*) = max{lD(A4)} = 0.757 which selects that terms of fuzzy dominance method explained by Bitarafan and Ataei
square set stoping method as most preferable. The program’s out- (2004).
put by the UMMS in Yager’s method section is presented in After reaching a solution, the decision maker can perform the
Fig. 10. At this section, the UMMS finds an alternative solution in sensitivity analysis in a similar way as previously presented in Sec-
S. Alpay, M. Yavuz / Tunnelling and Underground Space Technology 24 (2009) 173–184 183
Fig. 10. The program’s output by the UMMS in Yager’s method section.
tion 6.2 for the AHP method so as to analyze the sensitivity of the Bellman, R.E., Zadeh, L.A., 1970. Decision making in a fuzzy environment. Manage.
Sci. 17, 141–164.
solution.
Bieniawski, Z.T., 1989. Engineering Rock Mass Classification. John Wiley & Sons,
New York. 237 p.
7. Conclusions Bitarafan, M.R., Ataei, M., 2004. Mining method selection by multiple criteria
decision making tools. J. S. Afr. Inst. Min. Metall. 104, 493–498.
Boshkov, S.H., Wright, F.D., 1973 (Basic and parametric criteria in the selection,
In this paper, a computer program for underground mining design and development of underground mining systems). SME Mining
method selection (UMMS) was developed to eliminate the difficul- Engineering Handbook. SME-AIME, New York.
Chen, C., Klein, C.M., 1997. An efficient approach to solving fuzzy MADM problems.
ties in taking into consideration many decision criteria simulta- Fuzzy Set Syst. 88, 51–67.
neously in the underground mining method selection process Elevli, B., Demirci, A., 2004. Multicriteria choice of ore transport system for an
and to guide the decision makers to select the optimal under- underground mine: application of PROMETHEE methods. J. S. Afr. Inst. Min.
Metall. 104, 251–256.
ground mining method.
Elevli, B., Demirci, A., Dayi, O., 2002. Underground haulage selection: shaft or ramp
In the study, two methods, AHP and FMADM, and two AHP for a small-scale underground mine. J. S. Afr. Inst. Min. Metall. 102, 255–
models are proposed by the UMMS. Using the first AHP model, it 260.
Hartman, H.L., 1987. Introductory Mining Engineering. John Wiley, New Jersey.
is possible to analyze an underground mining method selection
Hartman, H.L., Mutmansky, J.M., 2002. Introductory Mining Engineering. John
problem with respect to the total of 36 criteria clustered in six Wiley, New Jersey.
main groups. This also makes possible to increase the consistency Kahriman, A., Ceylanoğlu, A., Demirci, A., Arpaz, E., Görgülü, K., 1996. Selection of
of decisions made for selection of the optimal mining method. On optimum underground mining method for Eskisehir-Karaburun chromite ore
(in Turkish). In: Proceeding of the 3rd National Rock Mechanics Symposium, pp.
the other hand, if the time is restricted and/or the six main group 47–60.
criteria without their sub-criteria are enough to make a decision, Karadogan, A., Bascetin, A., Kahriman, A., Gorgun, S., 2001. A new approach in
the second AHP model and/or the FMADM method can be used selection of underground mining method. In: Proceeding of the International
Conference-Modern Management of Mine Producing, Geology and Environment
as the alternative ways. Protection, pp. 171–183.
An important support of the UMMS to the decision making pro- Kazakidis, V.N., Mayer, Z., Scoble, M.J., 2004. Decision making using the analytic
cess is that the decision makers can perform the sensitivity analy- hierarchy process in mining engineering. Trans. Inst. Min. Metall. A 113, A30–
A42.
sis after reaching a solution. In this way, the decision makers can Kesimal, A., Bascetin, A., 2002. Application of fuzzy multiple attribute decision
observe how the proposed solution varies at a specific percentage making in mining operations. Miner. Resour. Eng. 11, 59–72.
change (increase and decrease) in the weights of the main criteria. Laubscher, D.H., 1981. Selection of Mass Underground Mining Methods. Design and
Operation of Caving and Sublevel Stoping Mines. SME-AIME, New York.
Miller-Tait, L., Panalkis, R., Poulin, R., 1995. UBC mining method selection. In:
References Proceeding of the Mine Planning and Equipment Selection Symposium, pp.
163–168.
Acaroglu, O., Ergin, H., Eskikaya, S., 2006a. Analytical hierarchy process for selection Morrison, R.G.K., 1976. AW Philosophy of Ground Control. McGill University,
of roadheaders. J. S. Afr. Inst. Min. Metall. 106, 569–575. Montreal, Canada.
Acaroglu, O., Feridunoglu, C., Tumac, D., 2006b. Selection of roadheaders by fuzzy Nicholas, D.E., 1981. Method Selection – A Numerical Approach. Design and
multiple attribute decision making method. Trans. Inst. Min. Metall. A 115, Operation of Caving and Sublevel Stoping Mines. SME-AIME, New York.
A91–A98. Riberio, R.A., 1996. Fuzzy multiple attribute decision making: a review and new
Ataei, M., 2005. Multicriteria selection for alumina-cement plant location in East- preference elicitation techniques. Fuzzy Set Syst. 78, 155–181.
Azerbaijan province of Iran. J. S. Afr. Inst. Min. Metall. 105, 507–514. Saaty, T.L., 1980. The Analytic Hierarchy Process. McGraw-Hill, Pittsburgh.
Bascetin, A., Kesimal, A., 1999. The study of a fuzzy set theory for the selection of an Saaty, T.L., 2000. Fundamentals of Decision Making and Priority Theory with the
optimum coal transportation system from pit to the power plant. Int. J. Surf. Analytic Hierarchy Process. RWS, Pittsburgh.
Min. Reclam. Environ. 13, 97–101. Samanta, B., Sarkar, B., Murherjee, S.K., 2002. Selection of opencast mining
Bascetin, A., Oztas, O., Kanli, A.I., 2006. EQS: a computer software using fuzzy logic equipment by a multi-criteria decision-making process. Trans. Inst. Min.
for equipment selection in mining engineering. J. S. Afr. Inst. Min. Metall. 106, Metall. A. 111, A136–A142.
63–70.
184 S. Alpay, M. Yavuz / Tunnelling and Underground Space Technology 24 (2009) 173–184
Triantaphyllou, E., Sánchez, A., 1997. A sensitivity analysis approach for some Yavuz, M., Alpay, S., 2007b. EQS: a computer software using fuzzy logic for
deterministic multi-criteria decision making methods. Decision Sci. 28, 151– equipment selection in mining engineering by A. Basçetin, O. Öztasß, and A.I.
194. Kanli in the Journal of SAIMM, vol. 106, No. 1, pp. 63–70. J. S. Afr. Inst. Min.
Yager, R.R., 1978. Fuzzy decision making including unequal objectives. Fuzzy Set Metall. 107, 138.
Syst. 1, 87–95. Yavuz, M., Iphar, M., Once, G., 2008. The optimum support design selection by using
Yavuz, M., 2007. Analytical hierarchy process for selection of roadheaders, by O. AHP method for the main haulage road in WLC Tuncbilek colliery. Tunn.
Acaroglu, H. Ergin, and S. Eskikaya in the Journal of SAIMM, vol. 106, No. 8, pp. Undergr. Space Technol. 23, 111–119.
569–575. J. S. Afr. Inst. Min. Metall. 107, 138. Zadeh, L.A., 1973. Outline of a new approach to the analysis of complex systems and
Yavuz, M., Alpay, S., 2007a. Mining method selection by multiple criteria decision decision process. IEEE Trans. 3, 28–44.
making tool by M.R. Bitarafan and M. Ataei in the Journal of SAIMM, vol. 104,
No. 9, pp. 493–498. J. S. Afr. Inst. Min. Metall. 107, 137.