Supplier selection based on two-phased fuzzy decision making / Fairuz Shohaimay...[et. al.]

Supplier selection depends on human evaluation which is subjective and vague in nature. Fuzzy approach is deemed appropriate to measure these uncertainties in the decision making process, rather than using real or crisp values. Predominant in many studies on fuzzy decision making, fixed triangular o...

Full description

Saved in:
Bibliographic Details
Main Authors: Shohaimay, Fairuz, Ramli, Nazirah, Mohamed, Siti Rosiah, Mohd, Ainun Hafizah
Format: Article
Language:English
Published: Universiti Teknologi MARA Cawangan Pahang 2013
Subjects:
Online Access:http://ir.uitm.edu.my/id/eprint/29602/1/29602.pdf
http://ir.uitm.edu.my/id/eprint/29602/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Supplier selection depends on human evaluation which is subjective and vague in nature. Fuzzy approach is deemed appropriate to measure these uncertainties in the decision making process, rather than using real or crisp values. Predominant in many studies on fuzzy decision making, fixed triangular or trapezoidal fuzzy numbers with symmetric spread from the literature were incorporated. However, these fuzzy numbers do not explain the actual respondents’ opinions which will affect the overall decision making process. Therefore, fuzzy numbers based on respondents should be developed beforehand to be integrated into the existing fuzzy decision making tool. This paper aims to develop triangular fuzzy numbers based on respondents’ opinions. These fuzzy numbers were adopted into a fuzzy evaluation method used in a supplier selection problem. The ranking results were analyzed using three different groups of fuzzy numbers. It was found that the linguistic terms for all three groups are not symmetric with the largest difference in spread that occurs for G2. There is also a variation in ranking of sub-criterion “Background of Supplier” in G2. Future studies in fuzzy decision making should include fuzzy numbers built based on respondents as they provide more reliable outcomes.