A Novel Decomposed Pythagorean Fuzzy CODAS Framework for Precision Customer Segmentation in Complex Market Environments

Abstract

Segmentation of customers has become an essential aspect in market analysis as it allows the organizations to formulate strategies according to different customer behaviour. These processes are further enhanced with advanced data mining techniques such as those used to uncover hidden patterns and take into account the heterogeneity of customers, which result in better and more data-driven decisions. To evaluate the quality of the segmentation-in particular for accuracy and practical usefulness of the solution, we take into account the performance measures Silhouette Score and Davies–Bouldin Index for accuracy and cohesion of the solution and Market Response. The proposed framework integrates the strong data mining procedures with New Disaggregated Pythagorean Fuzzy Sets based CODAS (NDPFS–CODAS) technique to increase the flexibility and discriminative power as well as the clearness of resultant segmentation. On the basis of behavioral, demographic and psychographic criteria, the model incorporates a multidimensional uncertainty. Case study confirms that NDPFS-CODAS enhances the accuracy and robustness of conventional fuzzy technique. These findings confirm its usefulness in target marketing, new-product development and customer and product in section marketing strategy, and give guidance for strategic decisions to offer the product packaging with maximum optimization.