Mining project portfolio selection (PPS) is a complex, capital-intensive decision-making challenge, often exacerbated by conflicting criteria and high uncertainty. To address this, this study introduces an integrated multi-criteria decision-making (MCDM) framework using trapezoidal Interval Type-2 Fuzzy Sets (TrIT2 FSs) to effectively capture the vagueness inherent in expert judgments. In this approach, we first utilize TrIT2 fuzzy numbers to quantify linguistic evaluations from decision-makers, followed by the entropy method to derive objective criteria weights. The MARCOS method is then employed to rank mining projects by evaluating their utility against ideal and anti-ideal solutions. The framework’s efficacy is demonstrated through a real-world case study of major mining projects in Iran, assessed by a panel of deputy managers. The results identify Golgohar (A5) as the top-performing project, while Zarshuran (A10) ranks lowest. Comparative analysis with TrIT2F-TOPSIS and TrIT2F-COPRAS confirms the stability and reliability of our rankings. Furthermore, sensitivity analyses across various strategic scenarios validate the robustness of the proposed model. Ultimately, the TrIT2F-Entropy-MARCOS framework offers a powerful, interpretable decision-support tool for strategic investment in the mining sector.