Consensus-reaching process in multi-stage large-scale group decision-making based on social network analysis: Exploring the implication of herding behavior Sun, Xinlei Zhu, Jianjun Wang, Jiepeng Pérez Gálvez, Ignacio Javier Cabrerizo Lorite, Francisco Javier Multi-stage large-scale group decision-making refers to a decision-making system with a large number of democratic evaluations provided by decision-makers and a multi-stage dynamic decision-making process. This method can provide the evaluation values of different decision-makers at heterogeneous stages, collect dynamic and complex decision-making information, and obtain objective decision-making results. Here, we consider large-scale group decision-making in the context of social networks, and develop a consensus-reaching process based on herding behavior. Decision-makers are identified based on the characteristics of information gradient dissemination, risk aversion, and authority obedience, and divided into different groups by how much herding behavior they exhibit. Guided by the referenced preferences obtained from a recommendation mechanism, the modification model considering the adjustment willingness and minimum adjustment cost, optimizes the preferences of low-consensus decision-makers in the strong and no herding groups. Meanwhile, the punishment model, which seeks the minimum degree of group consensus that satisfies the consensus threshold condition, optimizes the weight of the low-consensus decision-makers in the weak herding group. Finally, we present an illustrative example of emergency generation selection during the Spanish energy crisis to verify the rationality and soundness of the proposed multi-stage large-scale group decision-making approach. 2026-01-09T10:58:28Z 2026-01-09T10:58:28Z 2024-04 journal article Sun, X., Zhu, J., Wang, J., Pérez-Gálvez, I. J., & Cabrerizo, F. J. (2024). Consensus-reaching process in multi-stage large-scale group decision-making based on social network analysis: Exploring the implication of herding behavior. Information Fusion, 104, 102184. https://hdl.handle.net/10481/109381 10.1016/j.inffus.2023.102184 eng http://creativecommons.org/licenses/by-nc-nd/4.0/ open access Attribution-NonCommercial-NoDerivatives 4.0 Internacional Elsevier