In the present study, a distance-based fuzzy MCDM approach has been employed. The approach is based on the
distance from the ideal and non-ideal solutions. This was initially proposed by Karsak (2002). The novel feature of
this approach is that it can handle problems that are having both crisp and fuzzy data. This approach is interrelated
with technique for order preference by similarity to ideal solution (TOPSIS) (Hwang and Yoon 1981).
The traditional TOPSIS approach uses the Euclidean norm to normalise the original attribute values, and the
Euclidean distance to calculate each alternative’s distance from the ideal and anti-ideal solutions. The normalisation
and distance functions form the key components of the TOPSIS approach. Normalisation is performed to make the
criterion values unit-free and comparable. Besides the cumbersome computations required for applying the square
root operator to fuzzy data, a drawback of applying the Euclidean norm is that the minimum and the maximum
values of the normalised scale are not equal for each criterion, which results in difficulties in making inter-criterion
comparisons (Karsak 2002). In the present approach, some changes have been made to make it easier for
computation.
In the present MCDM approach, a linear scale transformation for normalisation, with an interval of 0 to 1, has
been employed. Therefore, there is no need to obtain the square root of fuzzy data, and it can address the problem
with crisp data, triangular fuzzy numbers, and linguistic terms simultaneously. The methodology of the proposed
approach is presented in the next section.
In the present study, a distance-based fuzzy MCDM approach has been employed. The approach is based on the
distance from the ideal and non-ideal solutions. This was initially proposed by Karsak (2002). The novel feature of
this approach is that it can handle problems that are having both crisp and fuzzy data. This approach is interrelated
with technique for order preference by similarity to ideal solution (TOPSIS) (Hwang and Yoon 1981).
The traditional TOPSIS approach uses the Euclidean norm to normalise the original attribute values, and the
Euclidean distance to calculate each alternative’s distance from the ideal and anti-ideal solutions. The normalisation
and distance functions form the key components of the TOPSIS approach. Normalisation is performed to make the
criterion values unit-free and comparable. Besides the cumbersome computations required for applying the square
root operator to fuzzy data, a drawback of applying the Euclidean norm is that the minimum and the maximum
values of the normalised scale are not equal for each criterion, which results in difficulties in making inter-criterion
comparisons (Karsak 2002). In the present approach, some changes have been made to make it easier for
computation.
In the present MCDM approach, a linear scale transformation for normalisation, with an interval of 0 to 1, has
been employed. Therefore, there is no need to obtain the square root of fuzzy data, and it can address the problem
with crisp data, triangular fuzzy numbers, and linguistic terms simultaneously. The methodology of the proposed
approach is presented in the next section.
การแปล กรุณารอสักครู่..