Developing RFM Model For Customer Segmentation Based On Fuzzy C.Means And Weighted Interval Valued Dual Hesitant Fuzzy Sets

Abstract

Marketing strategies have always been one of the top priorities for companies due to the effect on their revenue. In order to build a suitable marketing strategy, businesses need to have a clear understanding of their customers, and that proper customer segmentation is essential. Customer segmentation is one of the powerful tools for gaining an edge in a competitive environment. This thesis aims to make sufficient use of transaction data for identifying different types of customers by providing a new two-phase methodology for segmenting customers based on the RFM model. Phase 1 implements fuzzy clustering algorithms to segment the customers, where phase 2 focuses on ranking customers using weighted interval-valued dual hesitant fuzzy sets. This method classifies customers so that customers in each group have similar characteristics, and then assesses the importance of each customer group. A numerical application was provided to solve a real-life problem and thus the utilization of the proposed approach is proven. The results obtained from the model can be used to develop object-oriented marketing strategies or to develop customer relationship management campaigns.