Supplier Quality Evaluation Using a Fuzzy Multi Criteria Decision Making Approach - Complex System Modelling and Control Through Intelligent Soft Computations - page 209

Table 11 Closeness
coef cient (CCi) i ) for
alternatives

Alternatives
A1 A2 A3

d i 85.057 80.439 88.924

d i 101.802 104.824 101.616

CC i 0.455 0.434 0.467

The last row of Table 10 presents the distances d i and d i

for the three alter-

natives. Now, using distances d i and d i

(Eq. 14 ), we compute the closeness

coef cient (CCi) i ) of the three alternatives. For example, for alternative A1, the

closeness coef cient is given by:

CC i ¼ d i =ð d i þ d i Þ ¼

85 : 057 =ð 85 : 057 þ 101 : 802 Þ ¼ 0 : 455

Likewise, CCi i for the other two alternatives are computed. The final results are

shown in Table 11 .
By comparing the CCi i values of the three alternatives (Table 12 ), we nd that

A3 (0.467) > A1 (0.455) > A1(0.434). Therefore, A3 is selected as the supplier with
highest quality and is recommended for procurement of materials by the
organization.

4.2 Sensitivity Analysis

To investigate the impact of criteria weights (denoted by W Ci for criteria Ci where
i =1,2, ,n) on the selection of highest quality suppliers, we conducted the

sensitivity analysis. 23 experiments were conducted. In the

…

rst five experiments

(#1 5), weights of all criteria are set equal to (1, 1, 3), (1, 3, 5), (3, 5, 7), (5, 7, 9)

and (7, 9, 9) respectively. In experiments #6

-

22, the weight of each criteria is set as
highest (7, 9, 9) one by one and the remaining criteria are set to the lowest value (1,
1, 3). The goal is to see which criteria is most important in in

-

uencing the decision
making process. For example, in experiment #6, the criteria C1 has the highest
weight = (7, 9, 9) whereas the remaining criteria have weight = (1, 1, 3). In
experiment #23, the weight of the cost category criteria (C8) is set as lowest = (1, 1,
3), whereas the other criteria are set to highest weights = (7, 9, 9). The contrary of
experiment #23 is experiment #16 where weight of C8 = (7, 9, 9) and weight of
remaining criteria = (1, 1, 3). The details of the 23 experiments and their results are
presented in Table 12 .
Figure 1 presents the spider diagram for the results of the sensitivity analysis.
It can be seen from Table 12 and Fig. 1 that out of 23 experiments, alternative A3
(Supplier 3) has scored highest in 22/23 experiments followed by supplier A1

fl