A Probabilistic Linguistic Large-Group Emergency Decision-Making Method Based on the Louvain Algorithm and Group Pressure Model

To tackle preference conflicts and uncertainty in large-group emergency decision-making (LGEDM), this study proposes a probabilistic linguistic LGEDM method integrating the Louvain algorithm and group pressure model. First, expert weights are determined based on a social trust network, and the Louvain algorithm is employed for expert clustering, reducing the complexity of large-scale decision information. Second, a group pressure model is introduced to dynamically adjust expert preferences, enhancing consensus and decision consistency. Third, probabilistic linguistic term sets (PLTSs) are utilized to represent fuzzy and uncertain information, while attribute weights are determined by incorporating both subjective and objective factors, ensuring scientific rigor in decision-making. Finally, an improved TODIM (an acronym in Portuguese for Interactive and Multicriteria Decision-Making) method is adopted to account for the loss aversion behavior of decision-makers (DMs), enabling a more accurate characterization of psychological decision-making traits. The experimental results demonstrate that the proposed method outperforms existing approaches in terms of decision efficiency, group consensus, and result robustness, offering effective support for emergency decision-making in crisis situations.