Вы находитесь на странице: 1из 11

Journal of Intelligent Manufacturing, 13, 367±377, 2002

# 2002 Kluwer Academic Publishers. Manufactured in The Netherlands.

A fuzzy AHP approach to the determination of


importance weights of customer requirements in
quality function deployment
C . K . K W O N G and H . B A I
Department of Manufacturing Engineering, The Hong Kong Polytechnic University, Hung Hom,
Kowloon, Hong Kong
E-mail: mfckkong@inet.polyu.edu.hk

Received March and accepted November 2001

Quality function deployment (QFD) is an important tool in product planning that could contribute to
increase in customer satisfaction and shorten product design and development time. During the QFD
process, determination of the importance weights of customer requirements is a crucial and essential
step. The analytic hierarchy process (AHP) has been used in weighting the importance. However,
due to the vagueness and uncertainty existing in the importance attributed to judgement of customer
requirements, the crisp pairwise comparison in the conventional AHP seems to be insuf®cient and
imprecise to capture the degree of importance of customer requirements. In this paper, fuzzy number
is introduced in the pairwise comparison of AHP. An AHP based on fuzzy scales is proposed to
determine the importance weights of customer requirements. The new approach can improve the
imprecise ranking of customer requirements which is based on the conventional AHP. Finally, an
example of bicycle splashguard design is used to illustrate the proposed approach.

Keywords: Fuzzy AHP, quality function deployment, customer requirements, importance weights,
product design

1. Introduction of customers'', and hence to generate a set of


customer requirements. Determination of the impor-
Quality function deployment (QFD) is a management tance weights of customer requirements is a crucial
tool that provides a visual connective process to help and essential process of QFD as it could largely affect
teams focus on the customer requirements throughout the target value setting of engineering characteristics
the total product design and development cycle. QFD to be determined in the later stage. Various methods
is now being used for de®ning new products, as well as have been applied in this process. The simplest
for diagnosing and improving existing products. The method in prioritizing customer requirements is based
basic concept of QFD is to translate the desires of on point scoring scale, such as 1±5 and 1±10 (Grif®n
customers into appropriate product designs or engi- and Hauser, 1993). However, a substantial degree of
neering characteristics, and subsequently into parts human subjective judgement has to be involved in this
characteristics, process plans and production require- method. Lai et al. (1998) developed a group decision-
ments. It has been well documented that the use of QFD making technique to determine the importance
can reduce the product development time by 50%, and weights of customer requirements, in which the
start-up and engineering costs by 30% (Hauser and agreed criteria and individual criteria methods
Clausing, 1988). combine voting and linear programming techniques
As a customer-driven quality management tool, the to aggregate individual preferences into group
main characteristic of QFD is to recognize the ``voice consensus. Conjoint analysis method was attempted
368 Kwong and Bai

to determine the relative importance of customer requirements explicitly and determine the importance
requirements (Gustafsson and Gustafsson, 1994). The weights of customer requirements accurately.
methodology employs pairwise comparison of cus- In order to model this kind of uncertainty in human
tomer requirements to determine their relative preference, fuzzy sets could be incorporated with the
importance. Che et al. (1999) employed arti®cial pairwise comparison as an extension of AHP. The
neural network to determine the importance weights fuzzy AHP approach allows a more accurate
of customer requirements. To train the neural net- description of the decision-making process. The
work, a competitive assessment of the company and earliest work in fuzzy AHP appeared in Van
its competitors must be conducted in order to generate Laarhoven and Pedrycz (1983), which compared
the training data sets. Considering the vagueness and fuzzy ratios described by triangular membership
imprecision in the importance assessment of customer functions. Logarithmic least square was used to
requirements, Chan et al. (1999) directly converted derive the local fuzzy priorities. Later, using
the importance assessment from crisp values to fuzzy geometric mean, Buckley (1985) determined fuzzy
numbers, and then the importance weights of customer priorities of comparison, whose membership func-
requirements were calculated by using the entropy tions were trapezoidal. By modifying the Van
method. Prioritizing customer requirements could be Laarhoven and Pedrycz method, Boender et al.
viewed as a complex multi-criteria decision-making (1989) presented a more robust approach to the
problem. The analytic hierarchy process (AHP), a normalization of the local priorities.
multi-criteria decision making technique, was used in In this paper, a fuzzy AHP approach to the
weighing customer requirements (Lu et al., 1994). The determination of the importance weights of customer
integration of AHP with the determination of trade-off requirements for QFD is described. Firstly, the
weights for customer requirements has been proposed linguistic assessment on customer requirements is
(Akao, 1990; Aswad, 1989). Armacost et al. (1994) converted into triangular fuzzy numbers. These
adopted AHP to generate the importance weights triangular fuzzy numbers are used to build the
of customer requirements in a case study of comparison matrices of AHP based on pairwise
industrialized housing. Based on the customer comparison technique. The importance weights of
requirements and engineering requirements of a customer requirements can be calculated by applying
product collected from the QFD planning matrix, fuzzy AHP. At the end of this paper, an example of a
Zakarian and Kusiak (1999) applied AHP in the bicycle splashguard design is described to illustrate
determination of the importance measures for indivi- the fuzzy AHP approach to the determination of the
dual team members. importance weights of customer requirements.
In the conventional AHP, the pairwise comparisons
for each level with respect to the goal of customer
satisfaction are conducted using a nine-point scale. 2. Hierarchical structure for the development of
Each pairwise comparison represents an estimate of customer requirements
the priorities of the compared customer requirements.
The nine-point scale developed by Saaty (1980) AHP is particularly useful for evaluating complex
expresses preferences between options as equally, multi-attribute alternatives involving subjective cri-
moderately, strongly, very strongly, or extremely teria. The essential steps in the application of AHP
preferred. These preferences are translated into contains (1) decomposing a general decision problem
pairwise weights of 1, 3, 5, 7, and 9, respectively, in a hierarchical fashion into sub-problems that can be
with 2, 4, 6, and 8 as intermediate values. The easily comprehended and evaluated, (2) determining
pairwise comparison ratios are in crisp real numbers. the priorities of the elements at each level of the
However, customer requirements always contain decision hierarchy, and (3) synthesizing the priorities
ambiguity and multiplicity of meaning. The descrip- to determine the overall priorities of the decision
tions of customer requirements are usually linguistic alternatives.
and vague. Furthermore, it is also recognized that To apply AHP in prioritizing customer require-
human assessment on qualitative attributes is always ments, all customer requirements have to be
subjective and thus imprecise. Therefore, conven- structured into different hierarchical levels. Af®nity
tional AHP seems inadequate to capture customer diagram, tree diagram and cluster analysis can be used
Fuzzy AHP approach 369

Fig. 1. An example of a 3-level hierarchy for customer requirements.

for this purpose. Figure 1 shows an example of a F ˆ f…x; mF …x††; x [ Rg, where x takes its values on
three-level hierarchy for customer requirements. In the real line, R: ? 5 x 5 ‡ ? and mF …x† is a
the ®gure, the goal is ``customer satisfaction'', and continuous mapping from R to the closed interval
there are seven categories in the category level. All ‰0; 1Š. A triangular fuzzy number denoted as
customer requirements (attributes) are listed under M~ ˆ …a; b; c†, where a  b  c, has the following
relevant categories, which form the lowest level of the triangular-type membership function:
hierarchy. It is called attribute level. If there are a 8
large number of customer requirements, a four or >
> 0 x5a
<x a
more levels structure should be required. a  xb
mM~ …x† ˆ bc xa
>
> b  xc
:c b
0 x>c
3. Fuzzy representation of pairwise comparison
Alternatively, by de®ning the interval of con®dence
level a, the triangular fuzzy number can be
The hierarchy of customer requirements need to be
characterized as:
established before performing the pairwise compar-
ison of AHP. After constructing a hierarchy for Va [ ‰0; 1Š
customer requirements, the decision maker is asked ~ a ˆ ‰aa ; ca Š ˆ ‰…b
M a†a ‡ a; …c b†a ‡ cŠ
to compare the elements at a given level on a pairwise
basis to estimate their relative importance in relation Some main operations for positive fuzzy numbers
to the element at the immediate proceeding level. In described by the interval of con®dence (Kaufmann,
the conventional AHP, the pairwise comparison is 1991) are:
made using a ratio scale. A frequently used scale is the
nine-point scale which shows the participants' VmL ; mR ; nL ; nR [ R ‡ ; M ~ a ˆ ‰maL ; maR Š;
judgments or preferences among the options such as
N~a ˆ ‰naL ; naR Š; a [ ‰0; 1Š
equally, moderately, strongly, very strongly, or
extremely preferred. Even though the discrete scale ~ N~ ˆ ‰maL ‡ naL ; maR ‡ naR Š
M+
of 1±9 has the advantages of simplicity and easiness ~ N~ ˆ ‰maL naL ; maR naR Š
MY
for use, it does not take into account the uncertainty
associated with the mapping of one's perception (or ~ N~ ˆ ‰maL naL ; maR naR Š
M6
judgment) to a number. ~ N~ ˆ ‰maL =naR ; maR =naL Š
M
In this research, triangular fuzzy numbers, ~1 to ~9,
are used to represent subjective pairwise comparisons The triangular fuzzy numbers, ~1 to ~9, are utilized to
of customer requirements in order to capture the improve the conventional nine-point scaling scheme.
vagueness. A fuzzy number is a special fuzzy set In order to take the imprecision of human qualitative
370 Kwong and Bai
2 3
1 a~12 a~13  a~1…n 1† a~1n
6 a~21 1 a~23  a~2…n 1† a~2n 7
6 7
6 . .. .. .. .. .. 7
6 .. . . . . . 7
~
Aˆ6 7
6 .. .. .. .. .. 7
6 . . .  . . 7
6 7
4 a~…n 1†1 a~…n 1†2 a~…n 1†3  1 a~…n 1†n 5
a~n1 a~n2 a~n3  a~n…n 1† 1

where

1; iˆj
a~ij ˆ ~ 3; ~
1; ~ 7; ~
5; ~ 9 or ~
1 1
;~
3 1
;~
5 1
;~
7 1
;~
9 1
; i=j

Fig. 2. The membership functions of triangular fuzzy numbers Step 3: Solving fuzzy eigenvalues. A fuzzy eigen-
~ 3; ~
1; ~ 7; ~
5; ~ 9. value, ~l, is a fuzzy number solution to
~x ˆ ~l~
A~ x …1†

assessments into consideration, the ®ve triangular where A~ is a n 6 n fuzzy matrix containing fuzzy
fuzzy numbers are de®ned with the corresponding numbers a~ij and x~ is a non-zero n 6 1 fuzzy vector
membership functions as shown in Fig. 2. containing fuzzy numbers x~i .
To perform fuzzy multiplications and additions
using the interval arithmetic and a-cut, Equation 1 is
equivalent to
4. Algorithm of fuzzy AHP
‰aai1l xa1l ; aai1u xa1u Š+    +‰aainl xanl ; aainu xanu Š ˆ ‰lxail ; lxaiu Š
Saaty's AHP method is known as an eigenvector where
method. It indicates that the eigenvector corre-
sponding to the largest eigenvalue of the pairwise A~ ˆ ‰~
aij Š; x~t ˆ …~
x1 ; . . . ; x~n †;
comparisons matrix provides the relative priorities of  a  a
a~ij ˆ aijl ; aiju ; x~i ˆ ‰xail ; xaiu Š; ~la ˆ ‰lal ; lau Š …2†
a a
the factors, and preserves ordinal preferences among
the alternatives (Saaty, 1980). This means that if an for 0 5 a  1 and all i; j, where i ˆ 1; 2; . . . n; j ˆ 1; 2;
alternative is preferred to another, its eigenvector . . . ; n:
component is larger than that of the other. A vector of
Ä is
Degree of satisfaction for the judgment matrix A
weights obtained from the pairwise comparisons
matrix re¯ects the relative importance of the various estimated by the index of optimism m. The larger
factors. In the fuzzy AHP triangular fuzzy numbers value of the index m indicates the higher degree of
are utilized to improve the scaling scheme in the optimism. The index of optimism is a linear convex
judgment matrices, and interval arithmetic is used to combination (Lee, 1999) de®ned as:
solve the fuzzy eigenvector (Cheng and Mon, 1994). a^aij ˆ maaiju ‡ …1 m†aaijl ; Vm [ ‰0; 1Š …3†
The computational procedure of this methodology is
summarized as follows: While a is ®xed, the following matrix can be obtained
after setting the index of optimism, m, in order to
Step 1: Comparing the performance score. Triangular estimate the degree of satisfaction.
fuzzy numbers …~ 3; ~
1; ~ 7; ~
5; ~ 9† are used to indicate the 2 3
relative strength of each pair of elements in the same 1 a^a12    a^a1n
hierarchy. 6 a^a21 1    a^a2n 7
6 7
A~ ˆ 6 . .. .. .. 7 …4†
4 .. . . . 5
Step 2: Constructing the fuzzy comparison matrix. a^an1 a^an2    1
By using triangular fuzzy numbers, via pairwise
comparison, the fuzzy judgment matrix A Ä …aij † is The eigenvector is calculated by ®xing the m value and
constructed as shown below: identifying the maximal eigenvalue.
Fuzzy AHP approach 371
CR9 CR10 CR11 CR12 CR13 CR14
Step 4: Determining the total weights. By synthe- 2 ~ ~ ~ 1 ~ ~ 3
CR9 1 1 5 3 1 3 1
sizing the priorities over all levels, the overall 6~
CR10 61
1
1 ~
3 ~ 1
5 1 3 17
~ 7
importance weights of customer requirements are 6 1 7
CR11 6~5 ~
3 1 1 9 1 ~
~ 3 1 ~7 17
obtained by varying a value. S3: FCM3 ˆ 6
6 ~
7
CR12 6 3 ~
5 ~
9 1 ~
5 1 7
~ 7
6 1 7
CR13 4~1 1 ~
3 ~
5 1 1 ~
3 15
CR14 ~
3 ~
3 ~
7 ~
1 1 ~
3 1

5. An illustrative example " 16 CR#


CR 17
CR16 1 ~
1
S5: FCM4 ˆ
The design of a removable mountain bicycle CR17 ~
1 1 1
splashguard (Ullman, 1992) is used as an example
to illustrate the fuzzy AHP approach to the determina- 2 S1 S12 S3 1 3
S1 1 3 ~ ~
1
tion of the importance weights of customer 6~
C1: FCM5 ˆ S2 4 3 1 ~ 7
1 5
requirements.
S3 ~ ~ 1
1 1 1

" S4 S5 #
S4 1 ~
3
C2: FCM6 ˆ
S5 ~
3 1
1
5.1. Developing hierarchical structure
of customer requirements for bicycle " S6 S7 #
splashguard design S6 1 ~
1
C3: FCM7 ˆ
S7 ~
1 1
1
There are 19 customer requirements to be considered
in the design of a bicycle splashguard. They are 2 C1 C2 C3 3
classi®ed into three main categories and seven C1 1 ~
5 ~
7
6~ 1 ~ 7
subcategories. A four-level hierarchy of customer G: FCM8 ˆ C2 4 5 1 15
requirements for the splashguard design was con- C3 ~ 7 1 ~
1 1 1
structed as shown in Fig. 3.

5.3. Computing importance weights of customer


requirements
5.2. Constructing fuzzy comparison matrices The lower limit and upper limit of the fuzzy numbers
with respect to the a can be de®ned as follows by
Triangular fuzzy numbers, ~
1-~
9, are used to express the
applying Equation 2:
preference in the pairwise comparisons. By using
geometric means of the pairwise comparisons, the ~
1a ˆ ‰1; 3 2aŠ; ~ 3a ˆ ‰1 ‡ 2a; 5 2aŠ;
fuzzy comparison matrices (FCM) for each level can  
be obtained as follows: ~ 1 1
3a 1 ˆ ; ;
5 2a 1 ‡ 2a
 
CR1 CR2 CR3 CR4 CR5 CR6 ~ ~ 1 1 1
2 3 5a ˆ ‰3 ‡ 2a; 7 2aŠ; 5a ˆ ; ;
CR1 1 ~
3 ~
5 ~
7 ~
5 ~
7 7 2a 3 ‡ 2a
 
CR2 6~ 1 ~ ~ ~ ~7 1 1
63 1 1 3 1 3 7 ~
7a ˆ ‰5 ‡ 2a; 9 2aŠ; ~
7a 1 ˆ ; ;
6 1 7 9 2a 5 ‡ 2a
CR3 6~ ~ ~ ~ ~
37
S1: FCM1 ˆ 65 1 1 3 1 7  
CR4 6~ ~ ~ ~ 17 1 1
67 1 3 1
3 1 1 3 1 7 ~
9a ˆ ‰7 ‡ 2a; 11 2aŠ; ~ 9a 1 ˆ ; …5†
6 7 11 2a 7 ‡ 2a
CR5 4 ~ 5 ~
1 1
1 ~
3 1 ~
35
CR6 ~
7 1 ~
3 1 ~
3 1
1 ~
3 1 1
For example, let a ˆ 0.5 and m ˆ 0.5, and substitute
"CR7 CR#8 above expression into the fuzzy comparison matrices,
CR7 1 ~
1
S2: FCM2 ˆ FCM1 to FCM8 , all the a-cuts fuzzy comparison
CR8 ~ 1 1 1 matrices can be obtained as follows:
372 Kwong and Bai

Fig. 3. A hierarchy of customer requirements for bicycle splashguard design.

2 3
1 ‰1; 2Š ‰4; 6Š ‰6; 8Š ‰4; 6Š ‰6; 8Š
6 ‰1=4; 1=2Š 1 ‰1; 2Š ‰2; 4Š ‰1; 2Š ‰2; 4Š 7
6 7
6 7
6 ‰1=6; 1=4Š ‰1; 2Š 1 ‰2; 4Š ‰1; 2Š ‰2; 4Š 7
S1: FCMa1 ˆ 6
6 ‰1=8; 1=6Š
7
6 ‰1=4; 1=2Š ‰1=4; 1=2Š 1 ‰1=4; 1=2Š 1 7 7
6 7
4 ‰4; 6Š ‰1=2; 1Š 1 ‰2; 4Š 1 ‰2; 4Š 5
‰1=8; 1=6Š ‰1=4; 1=2Š ‰1=4; 1=2Š 1 ‰1=4; 1=2Š 1

 
1 ‰1; 2Š
S2: FCMa2 ˆ
‰1=2; 1Š 1
Fuzzy AHP approach 373
2 3
1 ‰1; 2Š ‰4; 6Š ‰1=4; 1=2Š ‰1; 2Š ‰1=4; 1=2Š
6 ‰1=2; 1Š 1 ‰2; 4Š ‰1=6; 1=4Š 1 ‰1=4; 1=2Š 7
6 7
6 7
6 ‰1=6; 1=4Š ‰1=4; 1=2Š 1 ‰1=10; 1=8Š ‰1=4; 1=2Š ‰1=8; 1=6Š 7
S3: FCM3 ˆ 6
a
6 ‰2; 4Š
7
6 ‰4; 6Š ‰8; 10Š 1 ‰4; 6Š ‰1; 2Š 7 7
6 7
4 ‰1=2; 1Š 1 ‰2; 4Š ‰1=6; 1=4Š 1 ‰1=4; 1=2Š 5
‰2; 4Š ‰2; 4Š ‰6; 8Š ‰1=2; 1Š ‰2; 4Š 1

 
1 ‰1; 2Š
S5: FCMa4 ˆ
‰1=2; 1Š 1

2 3
1 ‰1=4; 1=2Š ‰1=2; 1Š
6 7
C1: FCMa5 ˆ 4 ‰2; 4Š 1 ‰1; 2Š 5
‰1; 2Š ‰1=2; 1Š 1

 
1 ‰2; 4Š
C2: FCMa6 ˆ
‰1=4; 1=2Š 1

 
1 ‰1; 2Š
C3: FCMa7 ˆ
‰1=2; 1Š 1

2 3
1 ‰4; 6Š ‰6; 8Š
6 7
G: FCMa8 ˆ 4 ‰1=6; 1=4Š 1 ‰1; 2Š 5
‰1=8; 1=6Š ‰1=2; 1Š 1

Equation 3 and MATLAB package (Harman, 1997) eigenvectors of A can be calculated as follows by
are used to calculate eigenvectors for all comparison substituting the l1 into the equation, AX ˆ lX.
matrices, from which the importance weights of
individual customer requirements can be obtained.
T
For example, FCM0:55 can be obtained as shown below
X1 ˆ …0:6852; 0:2469; 0:6852†
after applying Equation 3.
2 3 After normalization, the importance weights of the
1:000 3:000 1:000
FCM5 ˆ 4 0:375 1:000 0:375 5
0:5 sub-categories of the customer requirements, S1 ; S2 ,
1:000 3:000 1:000 and S3 , can be determined as shown below.

Let FCM0:5
5 ˆ A. Eigenvalues of the matrix A can be
calculated as follows by solving the characteristic C1: ‰WS1 ; WS2 ; WS3 Š ˆ ‰0:4237; 0:1527; 0:4237Š
equation of A, det…A lI† ˆ 0.

l1 ˆ 3:081; l2 ˆ 0:081; l3 ˆ 0:000 Following the similar calculation, the importance


weights of C1 to C3 ; S4 to S7 and CR1 to CR19 can be
As the value of l1 is the largest, the corresponding determined as shown below.
374 Kwong and Bai

Table 1. Importance weights of customer requirements for bicycle splashguard design

Category Subcategory Attribute

Functional performance (0.7387) Attach/Detach (0.3130) Easy to attach (0.1278)


Easy to detach (0.0575)
Fast to attach (0.0423)
Fast to detach (0.0172)
Can attach when bike is dirty (0.0423)
Can detach when bike is dirty (0.0260)
Interface with bike (0.1128) Not mar (0.0661)
Not catch water, etc. (0.0467)
Structural integrity (0.3130) Not rattle (0.0489)
Not wobble (0.0331)
Not bend (0.0118)
Long life (0.1042)
Lightweight (0.0331)
Not release accidentally (0.0819)
Spatial constraints (0.1533) Fit (0.0635) Most bikes (0.0635)
Not interfere (0.0898) With lights and generator (0.0526)
With brakes (0.0372)
Appearance (0.1080) Shape (0.0447) Streamlined (0.0447)
Colour (0.0633) Popular colour (0.0633)

S1: ‰WCR1 ; WCR2 ; WCR3 ; WCR4 ; WCR5 ; WCR6 Š ˆ C3: ‰WS6 ; WS7 Š ˆ ‰0:4142; 0:5858Š
‰0:4082; 0:1386; 0:1351; 0:0551; 0:1351; 0:0829Š G: ‰WC1 ; WC2 ; WC3 Š ˆ ‰0:7387; 0:1533; 0:1080Š
S2: ‰WCR1 ; WCR8 Š ˆ ‰0:5858; 0:4142Š
S3: ‰WCR9 ; WCR10 ; WCR11 ; WCR12 ; WCR13 ; WCR14 Š ˆ Based on the above results, the total importance
weights of individual customer requirements can be
‰0:1564; 0:1058; 0:0376; 0:3330; 0:1058; 0:2616Š
calculated by using the following equations, and the
S5: ‰WCR16 ; WCR17 Š ˆ ‰0:5858; 0:4142 results are shown in Table 1.

TWS1 ˆ WC1 ? WS1 ; TWS2 ˆ WC1 ? WS2 ;


TWS3 ˆ WC1 ? WS3 ;
TWS4 ˆ WC2 ? WS4 ; TWS5 ˆ WC2 ? WS5 ;
TWS6 ˆ WC3 ? WS6 ;
TWS7 ˆ WC3 ? WS7 ; TWCR1 ˆ WC1 ? WS1 ? WCR1 ;
TWCR2 ˆ WC1 ? WS1 ? WCR2 ;
TWCR3 ˆ WC1 ? WS1 ? WCR3 ; TWCR5 ˆ WC1 ? WS1 ? WCR5 ;
TWCR4 ˆ WC1 ? WS1 ? WCR4 ;
TWCR6 ˆ WC1 ? WS1 ? WCR6 ; TWCR7 ˆ WC1 ? WS2 ? WCR7 ;
TWCR8 ˆ WC1 ? WS2 ? WCR8 ;
TWCR9 ˆ WC1 ? WS3 ? WCR9 ; TWCR10 ˆ WC1 ? WS3 ? WCR10 ;
TWCR11 ˆ WC1 ? WS3 ? WCR11 ;
TWCR12 ˆ WC1 ? WS3 ? WCR12 ; TWCR13 ˆ WC1 ? WS3 ? WCR13 ;
TWCR14 ˆ WC1 ? WS3 ? WCR14 ;
TWCR15 ˆ WC2 ? WS4 ; TWCR16 ˆ WC1 ? WS5 ? WCR16 ;
TWCR17 ˆ WC2 ? WS5 ? WCR17 ;
TWCR18 ˆ WC3 ? WS6 ; TWCR19 ˆ WC3 ? WS7

6. Discussions fuzzy number is the corresponding real crisp number.


The spread of the number is the estimation from the
In the pairwise comparisons of AHP, triangular fuzzy real crisp number. Equation 3 de®nes how the
numbers were introduced to improve the scaling estimated number, a^ij , reacts to the real crisp
scheme of Saaty's method. The central value of a number by adjusting the index of optimism, m. The
Fuzzy AHP approach 375

m indicates the degree of optimism, which could be The design of a bicycle splashguard was used as an
determined by design team. If the real crisp number is example to illustrate the application of fuzzy AHP
overestimated …m 4 0.5†, the value of a^ij is higher than method in the determination of the importance
the central value. If it is underestimated …m 5 0.5†, the weights of customer requirements for bicycle
value of a^ij is lower than the central value. splashguard design. The overall results show that the
By setting m value as 0.05, 0.5, and 0.95 combination of fuzzy decision making with AHP
respectively (re¯ecting the pessimistic, the moderate could become a useful tool for implementing QFD in
and the optimistic situations), three graphs as shown future research.
in Appendix A, B, and C were generated by using
MATALAB package with the a varying from 0 to 1.
From the graphs, mutual comparisons can be
performed from the most uncertain situation …a ˆ 0†
to the most certain situation …a ˆ 1†, from which
Acknowledgment
relative importance of the customer requirements
…CR1 *CR19 † can be realized. For example, from the
The work described in this paper was supported by a
three graphs, the importance weight of customer
grant from The Hong Kong Polytechnic University,
requirement CR7 is less than the one of the customer
Hong Kong (Project no. A-PC06).
requirement CR15 under the most uncertain compar-
ison …a ˆ 0† and highly optimistic situation
…m ˆ 0.95†. For the pessimistic situation …m ˆ 0.05†,
the importance weight of customer requirement CR7
is larger than the one of CR15 . For the moderate
situation …m ˆ 0.50†, the importance weights of Appendix A
the customer requirements CR7 and CR15 are very
close.

7. Conclusions

In this paper, a fuzzy AHP approach to the


determination of the importance weights of customer
requirements in QFD was presented. In the metho-
dology, triangular fuzzy numbers were introduced
into the conventional AHP in order to improve the
imprecise ranking of customer requirements. Using
fuzzy AHP in the determination of importance
weights of customer requirements has the following
two major advantages:
(1) Fuzzy numbers are preferable to extend the
range of a crisp comparison matrix of the conven-
tional AHP method, as human judgement in the
comparisons of customer requirements is fuzzy in
nature.
(2) Adoption of fuzzy numbers can allow design
team number of QFD to have freedom of estimation
regarding the overall customer satisfaction goal and
actual situations. Judgement can go from optimistic to
pessimistic.
376 Kwong and Bai

Appendix B References

Akao, Y. (1990) Quality Function Deployment: Integrating


Customer Requirements into Product Design,
Productivity Press, Cambridge, MA.
Armacost, R. T., Componation, P. J., Mullens, M. A. and
Swart, W. W. (1994) An AHP framework for
prioritizing custom requirements in QFD: An indus-
trialized housing application. IIE Transactions, 26(4),
72±79.
Aswad, A. (1989) Quality function deployment: A systems
approach, in Proceedings of the 1989 IIE Integrated
Systems Conference, Institute of Industrial Engineers,
Atlanta, GA, pp. 27±32.
Boender, C. G. E., de Grann, J. G., Lootsma, F. A. (1989)
Multicriteria decision analysis with fuzzy pairwise
comparison. Fuzzy Sets and Systems, 29, 133±143.
Buckley, J. J. (1985) Fuzzy hierarchical analysis. Fuzzy Sets
and Systems, 17, 233±247.
Chan, L. K., Kao, H. P., Ng, A. and Wu, M. L. (1999) Rating
the importance of customer needs in quality function
deployment by fuzzy and entropy methods.
International Journal of Production Research,
37(11), 2499±2518.
Che, A., Lin, Z. H. and Chen, K. N. (1999) Capturing weight
of voice of the customer using arti®cial neural network
in quality function deployment. Journal of Xi'an
Appendix C Jiaotong University, 33(5), 75±78.
Cheng, C. H. and Mon, D. L. (1994) Evaluating weapon
system by analytical hierarchy process based on fuzzy
scales. Fuzzy Sets and Systems, 63, 1±10.
Grif®n, K. and Hauser, J. R. (1993) The voice of the
customer. Marketing Science, 12(1), 1±27.
Gustafsson, A. and Gustafsson, N. (1994) Exceeding
customer expectations. The Sixth Symposium on
Quality Function Deployment, Novi, Michigan.
Harman, T. L. (1997) Advanced Engineering Mathematics
Using MATALAB, PWS Pub. Co., Boston.
Hauser, J. R. and Clausing, D. (1988) The house of quality.
Harvard Business Review, 66, 63±73.
Lai, Y. J., Ho, E. S. S. A. and Chang, S. I. (1998) Identify-
ing customer preferences in quality function deploy-
ment using group decision-making techniques, in
Usher, J., Roy, U. and Parsaei, H. (ed.), Integrated
Product and Process Development, Wiley, New York,
pp. 1±28.
Lee, A. R. (1999) Application of Modi®ed Fuzzy AHP
Method to Analyze Bolting Sequence of Structural
Joints, UMI Dissertation Services, A Bell & Howell
Company.
Lu, M. H., Madu, C. N., Kuei, C. and Dena Winokur (1994)
Integrating QFD, AHP and benchmarking in strategic
marketing. Journal of Business & Industrial
Marketing, 9(1), 41±50.
Fuzzy AHP approach 377

Saaty, T. L. (1980) The Analytic Hierarchy Process, extension of Saaty's priority theory. Fuzzy Sets and
McGraw-Hill, New York. Systems, 11, 229±241.
Ullman and David G. (1992) The Mechanical Design Zakarian, A. and Kusiak, A. (1999) Forming teams: An
Process, McGraw-Hill, New York. analytical approach. IIE Transactions on Design and
Van Laarhoven, P. J. M. and Pedrycz, W. (1983) A fuzzy Manufacturing, 31(1), 85±97.

Вам также может понравиться