Contractor selection in large-scale civil engineering projects is a critical challenge in project management due to the presence of conflicting evaluation criteria, information uncertainty, and divergent expert judgments. Conventional contractor selection approaches, which primarily emphasize the lowest bid price, are insufficient to simultaneously account for technical, managerial, and financial dimensions, often resulting in increased execution risks and compromised project performance. A review of the existing literature reveals that although fuzzy multi-criteria decision-making (MCDM) methods have been extensively applied to contractor selection, most studies fail to model expert disagreement and uncertainty in a structured and quantitative manner, particularly within group decision-making contexts typical of large-scale projects .To address this research gap, this study proposes a hybrid multi-criteria decision-making framework based on the Fuzzy Group Interval (FGI) approach for optimal contractor selection. In the proposed framework, group expert judgments are represented using interval fuzzy numbers, enabling the integrated handling of uncertainty in both criteria weighting and contractor performance evaluation. The applicability of the proposed framework is demonstrated through a real-world case study involving four contractor alternatives. The results indicate that “technical and operational capability” is the most influential criterion, with a final weight of 0.24. Among the evaluated contractors, Contractor P4 achieved the highest final score of 0.77 and was selected as the optimal alternative. Sensitivity analysis, conducted by applying ±10% variations to criteria weights, confirms the stability and robustness of the ranking results and supports the comparability and defensibility of the final decision.The findings demonstrate that the proposed FGI-based framework provides a transparent, robust, and effective decision-support tool for contractor selection in large-scale civil engineering projects, offering superior capability in managing uncertainty and reducing individual judgment bias compared to conventional MCDM methods.