Integrated fingerprinting and game-theoretic MCDA support chemistry-oriented prioritization of licorice germplasm beyond chemical similarity

Licorice ( Glycyrrhiza spp.) is a cultivated medicinal and aromatic crop, yet, robust germplasm evaluation remains challenging because chemical consistency, active-metabolite abundance and harvestable biomass are often assessed separately. To address this, we developed a decision-oriented framework integrating chromatographic fingerprinting, GLQFM and game-theoretic Multi-Criteria Decision Analysis. By establishing a Nash equilibrium between AHP-derived expert weights and entropy weights, with α = 0.457, the model balanced expert-informed biological relevance with data-driven discrimination. Across 18 accessions, our framework successfully rank the pharmacological-recognized G. uralensis as the preferred source of functional flavonoids. TOPSIS identified W-SB, W-YL, and W-HSK as the leading candidate accessions of G. uralensis ; this top-ranked group remained stable in TOPSIS-based Monte Carlo perturbation analysis and was consistently recovered by VIKOR and EDAS. Moreover, the method empirically rectified the evaluation bias of conventional fingerprinting: accession W-HSK, despite moderate fingerprint similarity (Rank 11), was prioritized as a leading candidate (Rank 3) due to its high bioactive content. These results show that fingerprint conformity and comprehensive chemical-yield performance provide complementary information for licorice germplasm evaluation. The proposed framework offers a practical and transferable approach for licorice germplasm screening and raw-material standardization in cultivated industrial crops.