Analyzing Customer Requirements Based on Text Mining via Spherical Fuzzy QFD

Abstract

Customer requirements (CRs) and their interpretation of the product is a crucial stage for a company’s existence in a competitive environment. QFD provides a structured methodology to define and rank the product technical requirements (TRs) related to CRs. Because the importance levels of CRs and the relationship between CRs and TRs have ambiguous information, it is profitable to use fuzzy numbers while implementing the QFD process. Nowadays, online reviews of users are an important source for collecting customer requirements. Text mining methods are helpful for extracting meaningful expressions on product attributes. This study proposed a methodology that integrates text mining methods for collecting CRs and spherical fuzzy numbers with QFD for ranking TRs. By the new proposed method, the CRs on smartwatches are extracted through Latent Dirichlet Allocation (LDA) method with specific suggestion words that is the first proposed in this study. Then, spherical fuzzy QFD is implemented to acquire weights of TRs. The smartwatch case is the first applied with text mining and QFD with spherical fuzzy numbers. © 2023 Elsevier B.V., All rights reserved.