Classical Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) and its fuzzy extensions frequently face two major challenges in complex decision environments: (i) handling multi-granular linguistic information arising from diverse expertise and cognitive styles, and (ii) adequately accounting for correlations among assessment criteria. Existing methods often address these challenges in isolation, resulting in information redundancy, biased weighting, and unstable ranking outcomes. To address these gaps, this paper proposes a novel hybrid multi-criteria decision-making (MCDM) framework that combines intuitionistic 2-tuple fuzzy linguistic (I2TFL) modeling with a correlation-aware enhancement of TOPSIS. The framework introduces four key methodological contributions: (i) a systematic normalization procedure that consolidates heterogeneous linguistic scales into a unified 2-tuple fuzzy domain; (ii) an entropy-based mechanism for objective determination of criterion weights; (iii) correlation modeling via enhanced Dice and Cosine similarity measures specifically developed for I2TFL representations; and (iv) integration of correlation adjustments into both the weighting and distance computations within TOPSIS. Theoretical analysis demonstrates desirable properties of the proposed similarity operators, and an extensive numerical case study illustrates the effectiveness of the framework. Comparative results show superior ranking discrimination, improved robustness against correlated criteria, and enhanced interpretability relative to classical TOPSIS and existing fuzzy extensions. These findings highlight the framework’s applicability for real-world decision-making problems characterized by uncertainty, linguistic heterogeneity, and interdependent criteria.
Mehmood et al. (Wed,) studied this question.