International Journal of Operations and Logistics Management www.absronline.org/journals
p-ISSN: 2310-4945; e-ISSN: 2309-8023
Volume: 3, Issue: 4, Pages: 351-371 (December 2014)
© Academy of Business & Scientific Research
*Corresponding author: Abbas Abbasi
Assistance Professor, Management Department, Shiraz University, Shiraz, Iran
E-Mail: aabbasi@shirazu.ac.ir
351
Supplier Selection Using Adaptive Neuro-Fuzzy Inference
System and Fuzzy Delphi
Abbas Abbasi
1
*, and Moloud Sadat Asgari
2
1. Assistance Professor, Management Department, Shiraz University, Shiraz, Iran
2. Master of Art in industrial Management, Management Department, Shiraz University, Shiraz, Iran
One of the most important functions of the purchasing management in every organization
is to settle on suitable providers. This decision is a sophisticated one since it requires
exercising great care in paying simultaneous attention to conflicting and heterogeneous
criteria. Therefore an appropriate method should be devised for supplier selection and
identifying evaluation criteria. This study tries to investigate and present such a method.
This study is carried out in two distinct phases. In the first phase, using previous
literature, the evaluative criteria for supplier selection are identified and using the fuzzy
Delphi method, the most important criteria for food industries are determined. In the
second phase, using ANFIS method, the best suppliers are chosen among 60 selected
suppliers. Ultimately, the data are analyzed by ANN method and the results obtained
from both methods are compared. The results indicate that the ANFIS method provides a
better performance compared to the ANN method regarding the supplier selection process
in such a way that it reaches higher scores in almost all of the criteria. This study uses the
proper method of fuzzy Delphi for determining the supplier evaluation criteria as well as
the most recent and exact ANFIS method for supplier selection.
Keywords: Supplier selection, adaptive neuro-fuzzy inference system (ANFIS), artificial
Neural Network (ANN), fuzzy Delphi method
INTRODUCTION
The growth and development of information and
communication tools in recent years have led the
world’s companies and organizations to expand
their markets all over the globe. Accordingly, the
traditional local monopoly markets have been
replaced by fully competitive global markets.
Hence, organizations have to use each opportunity
to optimize their competitive edge and commercial
activities. In order to realize this objective,
managers and scientists have agreed upon a
unique conclusion positing that in order for an
organization to survive the competitive
environment, it should work with supply chain
partners and in doing so it should improve the
performance of the entire chain (Lin et al., 2009).
The structure of a supply chain generally consists
of a combination of potential suppliers,
distributers, retailers and customers (Hugos, 2011).
In recent years, the researchers as well as
managers have found out that selecting the