The preference of DMs

Major strategic decisions often require input from multiple experts to ensure rational outcomes, yet experts frequently display herd behavior in group settings. We address thisproblem by propose a multi-stage group consensus model that incorporates updates to herding behavior states. Specifically, through the quantum order effects, the evolution of herding behavior is modelled, and the adjustment willingness of decision-maker is determined during this process. For low-consensus decision-makers, we calculate cognitive interference terms to construct a recommendation mechanism under the quantum interference effects. Then a modification model and a punishment model are developed to address the psychological and behavioral traits of both strong herding and no herding groups, by optimizing the preferences or decision weights to improve overall consensus. Finally, a case study on the selection of circulating water purification technology for a thermal power plant is used to demonstrate the superiority of the method proposed in the paper.