Spherical fuzzy extension of AHP‐ARAS methods integrated with modified k‐means clustering for logistics hub location problem

Abstract

AbstractLike most other industries, logistics services are currently encountering enormous transformation. Many companies worldwide are applying big data analytics to implement operational strategies and facilitate location selection. Logistics companies need to focus on now and explore some futures of logistics hub location problems. Creating a profitable logistic network is a crucial task for airline and postal services. To establish a cost‐effective network, transportation expenses should be decreased, and networks should be simplified. Hub location problem is born out of these needs. The concept of hub cumulates the flows and makes networks more reliable. In this article, a novel hub location selection approach is introduced in a group decision making (GDM) environment with uncertainty by integrating a modified weighted k‐means clustering algorithm with multi‐criteria decision‐making (MCDM) tools. The combined MCDM method integrates analytic hierarchy process (AHP) to measure criteria weights and Additive Ratio Assessment technique to measure the performance of hub location alternatives in a spherical fuzzy set (SFS) environment. The SFS has shown definite advantages in handling vagueness and uncertainty over crisp, fuzzy, or intuitionistic fuzzy sets to depict experts' evaluations with a richer structure, allowing for more representative decision making. Using Turkish logistics data, big data algorithm facilitates 15 possible locations, and these sites are ranked in order by the integrated GDM methodology. The validation of the proposed evaluation model is illustrated in an application of the network structure in Turkey. Finally, sensitivity and comparison evaluations are introduced to demonstrate the feasibility and effectiveness of the proposed approach.