The selection of an optimal supplier is a critical and open challenge in supply chain management. While experts' assessments significantly influence the supplier selection process, their subjective interactions can introduce unpredictable uncertainty. Existing methods have limitations in effectively representing and handling this uncertainty. The paper aims to address these challenges by proposing a novel approach that leverages q-rung orthopair fuzzy sets (q-ROFSs). The novelty of the proposed approach lies in its ability to accurately capture experts' preferences through the use of q-ROFSs, which offer membership and non-membership degrees, providing a broader expression space compared to conventional fuzzy sets. Additionally, it incorporates social network analysis (SNA) to effectively consider the trust relationships among experts. The proposed approach is divided into three stages. The first stage, presents a novel method to determine experts' weights by combining SNA, the Bayesian formula, and the maximum entropy principle. This approach allows for a more precise representation of varying levels of expertise and influence among experts, addressing the uncertainty arising from subjective interactions. The second stage introduces a hybrid weight determination method to determine criteria weights within the context of q-ROFSs. Entropy plays a crucial role in capturing fuzziness and uncertainty in q-ROFSs, while the projection measure simultaneously provides information about the distance and angle between alternatives. By employing both objective weights estimated using entropy and normalized projection measure and subjective weights derived using an aggregation operator and a score function, the presented approach achieves a comprehensive assessment of criteria importance. To incorporate both subjective and objective weights effectively, game theory is applied which allows us to align decision-making with both quantitative and qualitative aspects, making the method more versatile and applicable. The third stage redefines the traditional Combined Compromise Solution (CoCoSo) method using Bonferroni mean operators which captures interrelationships among arguments to be aggregated. Furthermore, in recognition of the importance of an expert risk preferences and psychological behaviors, we apply regret theory, ensuring that the chosen solutions align more effectively with their underlying preferences and aspirations. The applicability and effectiveness of the proposed approach are demonstrated through a numerical example of green supplier selection. The comparative analysis illustrates the practicality and real-world relevance while the sensitivity analysis, confirms the stability and robustness of the proposed approach.