Journal of Uncertain Systems

Vol.2, No.4, pp.289-298, 2008

Online at: www.jus.org.uk

Using Fuzzy Delphi Method in Maintenance Strategy Selection

Problem
Azizollah Jafari 1,3, Mehdi Jafarian1,3,∗, Abalfazl Zareei 2, Farzad Zaerpour 2
1
Department of Industrial Engineering, University of Science and Culture, Tehran, Iran
2
Department of Industrial Engineering, International University of Science and Technology, Tehran, Iran
3
Technology Development Institute, Sharif university branch of ACECR*, Iran
Received 5 March 2008; Accepted 10 May 2008

Abstract

In this paper, a new approach that applies fuzzy Delphi method in Simple Additive Weighting (SAW) is proposed
to be used in the Maintenance Strategy Selection Problem (MSSP). This approach is based on representation of the
importance of goals and each strategy capabilities to satisfy each maintenance goals with fuzzy numbers. Fuzzy
Delphi method is applied for the assessment of the importance of each goal and capability of each maintenance
strategy, considering the expert’s opinion. This method considers both tangible and intangible goals dealing with the
selection problem. Fuzzy Delphi method generates an L-R fuzzy number that measures information about the nature
of opinions more adequately. Yager ranking method is used in transforming fuzzy numbers in the selection problem.
Finally, through a heuristic algorithm, the main steps of the proposed method are presented.
© 2008 World Academic Press, UK. All rights reserved.
Keywords: maintenance, maintenance strategy selection problem, fuzzy Delphi method, fuzzy numbers, SAW
method

1 Introduction and Literature Review

Maintenance costs constitute a major part of total operating costs of all manufacturing and production plants [16]. The
selection of inefficient maintenance strategy has consequences on direct maintenance costs in an organization.
Selection of most efficient maintenance strategy is an important problem that an organization is dealing with. A
proper maintenance strategy, applied to equipment will save money for the organization. Many of the goals dealing
with the selection of the best maintenance strategy for equipment in an organization are non-monetary or intangible,
which beside the monetary goals makes the selection problem more complex.
In the literature, there are few studies that have been done on the MSSP. Murthy and Asgharizade [2] proposed
an approach for decision making when the company out sources the maintenance. They used game theory to conduct
a decision when the customer (the receptionist of maintenance) wants to decide whether having a service contract or
not. Bertolini and Bevilacqua [3] used a combined AHP-GP model for maintenance selection policy problem and in a
case study used it for identifying the optimal maintenance policy for a set of centrifugal pumps operating in the
process and service plants of an Italian oil refinery. Al-Najjar and Alsyouf [4] consider the most efficient maintenance
approach as the one that is able to provide and utilize the required information about the changes in the failure causes
behavior. They used past data and technical analysis of processes machines and components to identify the criteria for
an MCDM problem. They used fuzzy Inference System (FIS) to assess the capability of each maintenance approach.
Finally utilizing SAW, the efficient maintenance approach was selected. Löfsten [20] proposed a model based on cost
analysis to choose between corrective or preventive maintenance. Sharma [21] proposed an approach based on fuzzy
linguistic modeling to select the most effective maintenance strategy for the components/parts associated with the
system. Luce [22] proposed a method to select the most economical maintenance method by utilizing Weibull law in

∗
Corresponding author. Email: [email protected] (M. Jafarian)
290 A. Jafari et al.: Using Fuzzy Delphi Method in Maintenance Strategy Selection Problem

order to represent reliability and evaluate maintenance costs. Bevilacqua and Barglia [7] used AHP coupled with a
sensitivity analysis for maintenance strategy selection in an Italian Oil refinery. Triantaphyllou et al. [8] proposed a
method to find the criticality of each criteria dealing with maintenance strategies in which deals with the simplifying
of the complex maintenance criteria. Almedia and Bohoris [9] present a review of some basic decision theory
concepts and discussed their applicability in the selection of maintenance strategies. Ivy and Nembhard [13] integrate
statistical quality control (SQC) and partially observable Markov decision processes (POMDPs) for the evaluation of
maintenance policies under conditions of limited information.
One of the crucial steps in many decision making process is accurate estimation of pertinent data relevant to the
problem. The MSSP which is a multi criteria decision making (MCDM) problem faces the problem in estimating the
related factors. To solve this problem, some approaches using fuzzy concepts have been proposed. Wang et al. [5]
used fuzzy AHP for the evaluation of maintenance strategies. They used triangular fuzzy number (TFN) in fuzzy
AHP to model the uncertainty in the selection process. Mechaefske [6] used a fuzzy linguistic approach for the MSSP.
In his approach, the organization firstly select its goals then by interviewing the managers and employs the
importance of each goal and the capability of each maintenance strategy to satisfy each goal is captured, then by
utilizing some equations in the fuzzy environment the optimum maintenance strategy will be selected. The problem of
his approach is that he did not consider the variety of opinions and limit the opinions to deterministic linguistic
variables.
It is obvious that an efficient selection process must have the ability of determining the best maintenance strategy
under any uncertainty level. The variety of maintenance criteria and their importance in an organization have many
effects on the selection process and it makes the problem complex to deal with all of the variety in the selection
process. In this paper, a new approach to the MSSP is proposed which can determine the best maintenance strategy by
considering the uncertainty level and also all the variety in maintenance criteria and their importance. Fuzzy Delphi
method is used in SAW shown in a heuristic algorithm for the estimation of the importance of goals and the capability
of each maintenance strategy to satisfy each maintenance goal. This approach can consider all of the goals (tangible
or intangible) dealing with the selection of maintenance strategies in an organization. Consider that fuzzy Delphi
method generates L-R fuzzy numbers, which can measure more adequate information about the nature of opinions.
SAW as a best known and very widely used method of Multiple Attribute Decision Making (MADM) is used in this
paper in order to conduct the decision.
This paper is organized as follows: Section 2 reviews some popular maintenance strategies and represents some
factors in which an organization maybe consider for the selection of its maintenance strategy. Section 3 presents the
proposed method. Finally, through a heuristic algorithm in section 4, the main steps of the proposed method are
presented.

2 Maintenance Strategies and Maintenance Goals

2.1 Alternative Maintenance Strategies

In the following, three alternative maintenance strategies are considered in this paper and also a review of
maintenance goals divided into two different aspects (tangible and intangible) are presented.
In the literature there are several categorizations of maintenance strategies (e.g. Wang et al. [5]; Mechaefske [6];
Bertolini and Bevilacqua [3]) based on their goals in decision making or identification. In this paper we considered
only three maintenance strategies for illustrating the proposed selection method. Note that the proposed approach has
the flexibility to consider as many strategies as the organization wanted to insert in the selection process.
Failure based maintenance (FBM) is the original maintenance strategy which indicate that an equipment would
be repaired each time it breaks down and no action is performed to detect or prevent the failure. In this case, the
maintenance costs are usually high but sometimes it is cost effective [10].
Journal of Uncertain Systems, Vol.2, No.4, pp.289-298, 2008
291

To reduce frequent and sudden failures, preventive maintenance (PM) is applied, regardless of the condition of
the equipment. This strategy is based on periodically planned actions performed on the equipment intended to prevent
unscheduled downtime and premature equipment damage; see Mann et al. [11], Al-Najjar [10] and Mobley [17].
Preventive maintenance is a widely used strategy and also called time based preventive maintenance [5].
Condition based maintenance (CBM) is based on using critical components to predict failure of the equipment.
The maintenance is carried out just before the failure occurs. The maintenance decision is made using measured data
which is generated from monitoring systems such as vibration monitoring, lubricating analysis, and ultrasonic testing
[10]. CBM is usually used for rotating and reciprocating machines in example centrifugal pumps, turbines and
compressors.
Note that each organization has its own strategies based on the equipment used, therefore it is expected that an
organization perform some specific maintenance strategies. In this approach, we use three alternative maintenance
strategies but it should be considered that the approach is flexible and we can add any specified maintenance
strategies to the selection process.

2.2 Factors Dealing with the Selection of the Maintenance Strategies

When an organization wanted to select the best maintenance strategy for equipment, firstly, its maintenance goals
which are taken for comparing criteria must be set. Wang et al. [5] divided the maintenance goals into four aspects: (1)
safety (personnel, facilities, and environment); (2) cost (hardware, software, personnel training); (3) added-value
(spare parts inventories, production loss, and fault identification); (4) feasibility (acceptance by labors, technique
reliability). Maintenance goals can also be considered in two different aspects: (1) tangible (2) intangible. Tangible
goals are measurable and can be estimated using different tools i.e. low maintenance cost, improved reliability.
Intangible goals are not measurable but can be estimated using expert tools i.e. acceptance by labors, enhance
competitiveness. Note that we can measure some intangible factors using special indicators. This approach has the
ability to find the most efficient maintenance strategy considering all of the factors at the same time without
consideration of their tangible or intangible characteristic.
Different organizations have different maintenance goals, therefore, in this approach; it is offered to specify the
maintenance goals by interviewing the maintenance staff and managers.

3 Proposed Method

The organization firstly specifies the alternative strategies which are chosen to be performed on the equipment and
then by interviewing the maintenance staff and managers, the maintenance goals will be defined. Then by applying
Fuzzy Delphi method the importance of each goal and the capability of each maintenance strategy to satisfy each goal
will be determined. Using Yager ranking method the fuzzy numbers associated with the importance of each goal and
the capability of each maintenance strategy to satisfy each goal will be transformed to crisp values and finally by
specializing SAW as probably best known and very widely used method of MADM, the best maintenance strategy
will be selected.
Different tools used in order to determine the best maintenance strategy are described as follows.

3.1 Fuzzy Delphi Method

The Delphi method is an iterative process to collect and distill the anonymous judgments of experts. Fuzzy Delphi
method uses a serious of data collection and analysis techniques interspersed with feedback. The Delphi method is
well suited as a research instrument when there is incomplete knowledge about a problem.
Fuzzy Delphi method which was proposed by Chang et al.[12] is employed in this paper. Chang et al. [12] used
292 A. Jafari et al.: Using Fuzzy Delphi Method in Maintenance Strategy Selection Problem

fuzzy statistics and the technique of conjugate gradient search to derive membership function of L-R fuzzy number
(Fig.1) for the fuzzy forecasts. The procedure of Fuzzy Delphi method is illustrated in Appendix. The category of
shapes of membership function derived in this paper is as follows (consider that d ≤ 1 ):
( p)

⎧1 − ((b ( p ) − x) /(b ( p ) − a ( p ) )) L
( p)

, a ( p ) ≤ x ≤ b( p )
⎪
⎪1 , b( p ) ≤ x ≤ c( p )
µ ( x) = ⎨
( p)
( p)
(1)
⎪1 − (( x − c ) /( d − c )) , c( p ) ≤ x ≤ d ( p )
( p) ( p) ( p) R

⎪
⎩0 , otherwise,
where P is the Pth survive question.

Figurer 1: Membership function of L-R fuzzy number

The survive questions in the MSSP are the importance of each goal ( µ j ( x) ) and the capability of each
I

maintenance strategy to satisfy each goal ( µ j ( x ) ).

Ci

After finding the membership function of the importance of jth goal ( µ j ( x) ) and also the membership function
I

of the capability of ith maintenance strategy to satisfy jth goal ( µ j ( x ) ), an approach based on a fuzzy transformation
Ci

methods will be used for the selection of the best maintenance strategy.

3.2 Yager Ranking Index

Yager [14] proposed a procedure for ordering fuzzy sets based on the concept of area compensation. Area
compensation possesses the properties of linearity. A ranking Index I ( p ) is calculated for the convex fuzzy number
p from its α-cut α  = [ P L , PU ] according to the following formula:
p α α

1
I ( p ) = ∫
1
( pαL + pαU ) dα (2)
0
2
which is the center of the mean value of p . Consider two fuzzy number p1 and p 2 , the equation I ( i
p1 ) ≥ I ( i
p2 )
implies that i
p1 ≥ i
p2 [14,15]. This index is very simple to apply.
Yager ranking index is used as a transformation method for the membership function of the importance of jth
goal ( µ j ( x) ) and also the membership function of the capability of ith maintenance strategy to satisfy jth goal
I

( µ j ( x ) ).
Ci
Journal of Uncertain Systems, Vol.2, No.4, pp.289-298, 2008
293

 j ) and I (C j ) , respectively.

The ranking index of µ j ( x) and µ j ( x ) will be denoted as I (G
I Ci
i

 j) = 1 1  L  j )U )dα
I (G ∫ 0
2
((G j )α + (G α , for j = 1,..., m. (3)

 j) = 1 1 i L  ij )U )dα
∫
i
I (C ((C j )α + (C α , for i = 1,..., n and j = 1,..., m. (4)
0
2

3.3 Simple Additive Weighting

In order to make decisions in the presence of multiple and conflicting criteria Multiple Attribute Decision Making
(MADM) is applied. In MADM, The alternatives have associated with them a level of the capability in satisfying the
attributes based on which the final decision is to be made. The final decision is conducted with the help of inter and
intra-attribute comparisons (see [18], [19]).
SAW is probably the best known and very widely used method of MADM. To each of the attributes in SAW, the
decision maker assigns importance weights which become the coefficients of the variables. The decision maker can
then obtain a total score for each alternative simply by multiplying the scale rating for each attribute value by the
importance weight assigned to the attribute and then summing these products over all attributes.
SAW method can be stated as follows. Note that the formula in SAW is rearranged to become proper to proposed
approach. Suppose the importance of each goal ( µ j ( x) ) are assigned a set of importance weights to the goals,
I

µ ( x) = { µ1 ( x) , µ 2 ( x) , ... µ n ( x) } and the capability of each maintenance strategy to satisfy each goal µ j ( x)
I I I I Ci

are also defined through utilizing fuzzy Delphi method. Then the most efficient maintenance strategy S* is selected
such that
n n
S * = {Si | max
i
∑ µ Cj ( x) × µ Ij ( x) / ∑ µ Ij ( x)}
j =1
i

j =1
(5)

4 Hypothetical Example
4.1 Conceptual Model of Proposed Method

As shown in Fig.2, the suggested approach consists of five main steps:

(1) Assessment of the importance of each goal using fuzzy Delphi method
(2) Assessment of the capability of each maintenance strategy to satisfy each maintenance goal using fuzzy
Delphi method
(3) Defuzzifying of the L-R fuzzy numbers G  j and C ij using transformation methods
(4) Rank ordering the maintenance strategies using SAW method
(5) Selecting the most efficient maintenance strategy
294 A. Jafari et al.: Using Fuzzy Delphi Method in Maintenance Strategy Selection Problem

Identification of goal Identification of

which organization set for maintenance strategies
its maintenance strategy applicable to the equipment

(2)Assessment of the capability of each

(1)Assessment of the importance of each goal maintenance strategy to satisfy each
using fuzzy Delphi method maintenance goal using fuzzy Delphi method

(3)Defuzzifying of the L-R fuzzy numbers

 j and C ij
G

(4)Rank ordering the maintenance strategies

using SAW method

(5)Selecting the most efficient maintenance

strategy

Figure 2: Conceptual model of the proposed approach

4.2 Heuristic Algorithm

A heuristic algorithm is presented based on the proposed approach for the selection of best maintenance strategy to be
performed on specific equipment. It is assumed that the availability of each maintenance strategy is constant. The
algorithm is presented in five steps:
Step 1: Organization specifies maintenance strategies wanted to apply to the equipment and the maintenance
goals for its maintenance strategy. Consider that the goals are based on the factors in which is important to the
organization. These goals maybe some or all of the following:
(1) Low maintenance cost (Tangible [ 6,7,5,13])
(2) Acceptance by labors (Intangible [6,5])
(3) Improved reliability (Tangible [6,1,5])
(4) Enhance Competitiveness (Intangible [6])
(5) High product quality(Tangible [6,7])
(6) Minimum inventories(Tangible [6,7,5])
The strategies that the organization selected are:
(1) Failure based maintenance (FBM)
(2) Preventive maintenance (PM)
(3) Condition based maintenance (CBM)
Step 2: Using fuzzy Delphi method by interviewing managers and employees of the maintenance department of
the organization as experts, the membership function of the importance of jth goal ( µ j ( x ) ) is derived (Table 1).
I
Journal of Uncertain Systems, Vol.2, No.4, pp.289-298, 2008
295

Table 1: Fuzzy weight of the importance of each goal

Goal Importance
L R a b c d
Low maintenance cost 2.3 4.5 0.6 0.8 0.8 0.1
Feasibility (acceptance by labors ) 5.3 2.1 0.3 0.4 0.4 0.5
Improved reliability 4.1 5.7 0.4 0.5 0.5 0.8
Low downtime length 1.2 4.3 0.6 0.74 0.74 0.91
Improved safety 2.5 4.1 0.3 0.71 0.71 0.8
High product quality 6.2 8.1 0.6 0.89 0.89 0.94
Minimum inventories 1.4 1.7 0.3 0.4 0.4 0.49

Note that the universe of discourse was U = [0,1] in which one is the highest degree of importance.
Step 3: Using fuzzy Delphi method, the membership function of the capability of ith maintenance strategy to
satisfy jth goal ( µ j ( x ) ) is derived (Table 2).
Ci

Table 2: Fuzzy weight of the capability of each maintenance strategy to satisfy each goal

Strategy
FBM PM CBM
Goal
L R A b c d L R A b c d L R A b c d
1 5.5 4.6 0.3 0.4 0.4 0.5 6.8 9.5 0.5 0.8 0.8 0.94 2.7 6.8 0.4 0.6 0.6 0.7
2 3.1 4.5 0.5 0.7 0.7 0.8 9.7 7.1 0.4 0.6 0.6 0.7 1.9 6.1 0.4 0.5 0.5 0.9
3 6.8 3.7 0.3 0.6 0.6 0.7 2.6 0.8 0.4 0.5 0.5 0.9 2.8 9.7 0.5 0.6 0.6 0.7
4 3.6 2.5 0.2 0.4 0.4 0.5 7.2 8.8 0.5 0.7 0.7 0.9 1.3 0.7 0.5 0.6 0.6 0.68
5 5.8 6.2 0.4 0.6 0.6 0.7 0.6 0.7 0.71 0.8 0.8 0.91 9.1 5.5 0.5 0.7 0.7 0.9
6 7.3 7.1 0.4 0.6 0.6 0.7 8.9 1.3 0.7 0.8 0.8 0.84 6.9 2.5 0.5 0.6 0.6 0.8
7 8.6 5 0.3 0.4 0.4 0.6 0.1 7.5 0.53 0.6 0.6 0.8 0.5 1.6 0.3 0.5 0.5 0.7

Step 4: Utilizing Yager ranking method, the membership functions defuzzified to crisp numbers. Crisp values of
membership functions of importance of each goal and the capability of each strategy to satisfy each maintenance goal
are illustrated in Tables 3 and 4.
A MATLAB ® is used to solve the algorithms of Yager ranking index of membership functions.
Step 5: Based on the SAW method, the values of S1 , S 2 and S3 for PM, CBM, and FBM, respectively, are
calculated and shown in Table 5. Based on the measured values, the maintenance strategies ranked and the best
maintenance strategy will be selected.
296 A. Jafari et al.: Using Fuzzy Delphi Method in Maintenance Strategy Selection Problem

Table 3: Crisp values of the importance of each goal using Yager ranking method

Goal Importance
Yager ranking method
Low maintenance cost 0.7879
Feasibility (acceptance by labors ) 0.4082
Improved reliability 0.5126
Low downtime length 0.7242
Improved safety 0.7397
High product quality 0.8726
Minimum inventories 0.4144

Table 4: Crisp values of the capability of maintenance strategies using Yager ranking method

Strategy
Goal FBM PM CBM
1 0.4012 0.7874 0.5794
2 0.6938 0.5968 0.5109
3 0.6020 0.5972 0.5915
4 0.3925 0.6980 0.6018
5 0.5922 0.8042 0.7055
6 0.5938 0.8254 0.6222
7 0.4115 0.7742 0.5718

Table 5: Final results

Maintenance strategy Si
Preventive maintenance (PM) 0.742569
Condition based maintenance (CBM) 0.606743
Failure based maintenance (FBM) 0.519974

4.3 Result Discussion

The value of Si declares the strategy’s tendency to capture the goals therefore each strategy which has the highest Si
value has the most power to capture the goals which the organization defined. Preventive maintenance is selected as
the best maintenance strategy in this case because of its highest preference value.
Note that to use fuzzy Delphi method, the maintenance staff and managers should participate because the more the
participants are aware of the problem, the more accurate the result would be.

5 Conclusion

The MSSP must be considered as an important management problem because of its effective roles in production and
manufacturing. The accuracy in the selection of efficient maintenance strategy for an equipment is based on
determining the right maintenance goals in the decision making process. The main problem in decision making
process is that there are some goals which are intangible. To deal with the right decision in the selection of best
maintenance strategy, in this paper, fuzzy Delphi method is applied. Yager ranking method is employed to transfer
the result of fuzzy Delphi method to crisp values. Hopefully, this approach can assist decision makers in the selection
Journal of Uncertain Systems, Vol.2, No.4, pp.289-298, 2008
297

of the most efficient maintenance strategy. Further research can be focused on applying fuzzy Delphi method to AHP
to measure the intangible criteria dealing with the selection of best maintenance strategy.

Appendix: Fuzzy Delphi Method Used in this Paper [12]

Step 1: Set the iteration counter k equal to one.
(i ) (i )
Step 2: A group of n experts is desired to give an interval-valued opinion [ qk , rk ] on each survey where i is the ith
expert and p is the pth survey question.
Step 3: For each survey item p compute a discrete membership function:
(1) Let U = [0,1] be the universe of discourse. Partition U into S contiguous interval:
I 1, I 2,..., Is .For each Is let xs denotes the midrange of Is to represent Is .
( p)
(2) Calculate ys as follow:
n
ys( p ) = ∑ δ s( i , p ) , (A1)
i =1
where
⎪⎧1 if xs ∈ [qk(i ) , rk(i ) ]( p )
δ (i, p )
s =⎨ (A2)
⎪⎩0 otherwise.
(3) The discrete membership function for survey p then can be characterized by
ys( p )
Y ( p ) ( xs ) = , s = 1,..., S . (A3)
y*( p )
where
y*( p ) = max { ys( p ) } . (A4)
s =1,..., S
Step 4: For each survey p, use the results obtained in Step 3 and the technique of the conjugate gradient search to
obtain the continuous mathematically explicit membership function. The fuzzy opinion for the survey can be
conveniently denoted as Ok
( p)
= (ak( p ) , bk( p ) , ck( p ) , d k( p ) ) with ak( p ) ≤ bk( p ) ≤ ck( p ) ≤ d k( p ) .
( p)
Step 5: Generate the feedback information for the next iteration by using the α-level cut of fuzzy opinion Ok :
[O ( p)
k α] = {x | µ ( p)
k ( x) ≥ α } = {l ( p)
k ( x), u ( p)
k ( x)}. (A5)
Step 6: Prepare data for stability testing of fuzzy Delphi process, by using the OM index for each item p:
1
OM (Ok( p ) ) = ∫ ϖ ( w)[ x1( w) × lk( p ) ( w) + x 2( w) × uk( p ) ( w)]dw , (A6)
ρ*

Where ϖ ( w) denotes the weighting measure of the w-level cuts of Ok

( p)
and x1( w) and x 2( w) the measures of
( p)
pessimism an optimism, respectively, of O k under w and also ρ * assumed to be greater than zero.
Step 7: If all the differences of the fuzzy opinions between two consecutive iterations for each item p appear to be
smaller than the given criterion δ , the process is complete and the final estimates are the fuzzy opinions obtained at
the last iteration; Otherwise increase k by one and return to Step 2.

References
[1] Chan, G.K., and S. Asgarpoor, Optimum maintenance policy with Markov process, Electronic Power Systems Research,
vol.76, pp.452-456, 2006.
[2] Murthy, D.N.P., and E. Asgharizade, Optimal decision making in a maintenance service operation, Eurpean Journal of
Operational Research, vol.116, pp.259-273, 1999.
[3] Bertolini, M., and M. Bevilacqua, A combined goal programming—AHP approach to maintenance selection problem,
Reliability Engineering & System Safety Roach to Maintenance Selection Problem, vol.91, no.7, pp.839-848, 2006.
298 A. Jafari et al.: Using Fuzzy Delphi Method in Maintenance Strategy Selection Problem

[4] Al-Najjar, B., I. Alsyouf, Selecting the most efficient maintenance approach using fuzzy multiple criteria decision making,
International Journal of Production Economics, vol.84, pp.85-100, 2003.
[5] Wang, L., J. Chu, and J. Wu, Selection of optimum maintenance strategies based on a fuzzy analytic hierarchy process,
International Journal of Production Economics, vol.107, pp.151-163, 2007.
[6] Mechaefske, C.K., Using linguistics to select optimum maintenance and condition monitoring strategies, Mechanical and
System and Processing, vol.17, no.2, pp.305-316, 2003.
[7] Bevilacqua, M., and M. Barglia, The analytic hierarchy process applied to maintenance strategy selection, Reliability
Engineering and System Safety, vol.70, pp.71-83, 2000.
[8] Triantaphyllou, E., B. Koalerchuk, et al.., Determining the most important criteria in maintenance decision making, Quality in
Maintenance Engineering, vol.3, no.1, pp.16-28, 1997.
[9] de Almedia, A.T., and G.A. Bohoris, Decision theory in maintenance decision making, Quality in Maintenance Engineering,
vol.1, no.1, pp.39-45, 1995.
[10] Al-Najjar, B., An approach for continous reduction in LLC when using integrated vibration-based maintenance: A case study,
COMADEM 99, Sunderland University, Sunderland, England, pp.159-169, 1999.
[11] Mann, L., A. Saxena, and G. Knapp, Statistical-based or condition-based maintenance?, Journal of Quality in Maintenance
Engineering, vol.1, no.1, pp.46-59, 1995.
[12] Chang, P. T., L.C. Huang, and H.J. Lin, The fuzzy Delphi via fuzzy statistics and membership function fitting and an
application to human resources, Fuzzy Sets and Systems, vol.112, pp.511-520, 2000.
[13] Ivy, J.S., and H.B. Nembhard, A modeling approach to maintenance decisions using statistical quality control and
organization, Quality and Reliability Engineering International, vol.21, pp.355-366, 2005.
[14] Yager, R.R., A procedure for ordering fuzzy subsets of the unit interval, Inform., vol.24, pp.143-161, 1981.
[15] Fortemps, P., and M. Roubens, Ranking and defuzzification methods based on area compensation, Fuzzy Sets System, vol.82,
pp.319-330, 1996.
[16] Mobley, R.K., An Introduction to Predictive Maintenance, 2th ed., Elsevier Science (USA), 2002.
[17] Mobley, R.K., Maintenance Fundamentals, 2th ed. Elsevier Inc, 2004.
[18] Yoon, K.P., and C.L. Hwang, Multiple Attribute Decision Making, an Introduction, Sage Publications, Thousand Ocks, 1995.
[19] Barba-Romero, S., J.C. Pomerol, , Decisiones Multicriterio. Fundamentos Teóricos y Utilización Práctica. Universidad de
Alcalá, Alcalá de Henares, 1997.
[20] Löfsten, H., Management of industrial maintenance – economic evaluation of maintenance policies, International Journal of
Operations & Production Management, vol.19, no.7, pp.71-73, 1999.
[21] Sharma, R.K., D. Kumar, and P. Kumar, FLM to select suitable maintenance strategy in process industries using MISO model,
Journal of Quality in Maintenance Engineering, vol.11, no.4, pp.359-374, 2005.
[22] Luce, S., Choice criteria in conditional preventive maintenance, Mechanical Systems and Signal Processing, vol.13, no.1,
pp.163-168, 1999.