A Data‐Driven Analysis of Multicriteria Decision‐Making for Supplier Selection: A Systematic Literature Review Using Topic Modeling and Logistic Regression

Supplier selection increasingly involves the evaluation of multiple conflicting criteria in data‐intensive organizational environments, driving the adoption of multicriteria decision‐making (MCDM) methods and Business Intelligence (BI) technologies. Despite these advances, the literature still lacks a systematic understanding of how these approaches are jointly applied and how decision‐related knowledge is structured and analytically integrated. This study addresses this gap through a data‐driven, mixed‐method approach based on methodological triangulation, combining a Systematic Literature Review (SLR), Latent Dirichlet Allocation (LDA), and binomial logistic regression. The results indicate that quality, cost, and delivery are the most recurrent supplier selection criteria within the reviewed AHP–TOPSIS literature. LDA identifies underlying thematic structures associated with criteria, methods, and application contexts, highlighting sectoral heterogeneity and the limited explicit representation of BI architectures in the literature. The logistic regression model shows that technological integration (Integrated_IT) is positively and statistically associated with methodological extension, increasing the likelihood of incorporating additional approaches beyond AHP–TOPSIS. These findings indicate that BI operates not only as an operational support mechanism but also as an enabler of analytical capability development and methodological hybridization, revealing a gap between technological use and its formal representation in decision models. The study contributes to knowledge and process management by providing quantitative evidence on this relationship and by advancing an integrated analytical approach for investigating complex decision‐making phenomena in data‐driven organizational contexts.