A fuzzy logic mindset agency theory decision support model for diagnosing consumer resistance to eco-friendly cosmetics

Eco-friendly cosmetics receive generally positive attitudes but continue to underperform commercially, indicating the presence of latent consumer resistance that firms struggle to diagnose. This study introduces a fuzzy logic Decision Support System grounded in Mindset Agency Theory (MAT) that quantifies resistance and supports scenario-based interventions on sensory experience, value perceptions, and ethical cues. We analyse 33 public survey reports (~110,000 respondents), apply TF–IDF vectorisation and a multi-step k-means model selection procedure, obtaining six coherent lexical clusters validated through multi-index optimisation and bootstrap stability (Jaccard, ARI). Eight lexicon-based MAT profiles are computed for each document and aggregated at cluster level, producing graded cluster to archetype mappings rather than rigid categorical assignments. A block PCA procedure reduces the eight profiles to the three MAT agencies—Affect, Cognition, Spirit—which serve as inputs to a type 1 Mamdani Fuzzy Inference System generating a continuous Resistance score. Representation agnostic triangulation using embedding based clustering confirms that detected clusters capture stable semantic patterns rather than TF–IDF artefacts (ARI = 0.253; NMI = 0.522). The resulting EcoResistanceFIS provides an interpretable and scenario ready tool connecting mindset configurations to actionable managerial levers, enabling targeted strategies to reduce consumer resistance in sustainable beauty markets.