In this study, we propose a statistical method incorporating decision risk and risk attitude into large group decision-making. The decision-making groups are divided into subgroups based on their attitudes to risk, and the evaluation information for the decision-makers within the same subgroup is combined to form the sample dataset. Next, the internal decision risk levels of all subgroups are measured using sample standard deviations and reduced through a feedback mechanism. Significance testing is used to determine the criterion weights and to measure the external decision risk levels of subgroups. The internal and external decision risk levels are then combined to yield the subgroup weights. Confidence interval is used to transform the sample data into interval numbers, which are then aggregated and analyzed to yield the decision-making results. Meanwhile, risk attitudes are taken into account throughout the decision-making process by various means. A case study and comparison analyses, along with sensitivity analyses, are used to illustrate the feasibility and rationality of the proposed method. Our experiments suggest that decision risk and risk attitude matter in large group decision-making.

在这项研究中，我们提出了一种将决策风险和风险态度纳入大集团决策的统计方法。根据决策者对风险的态度将其划分为子组，并将同一子组中决策者的评估信息合并以形成样本数据集。接下来，使用样本标准偏差测量所有子组的内部决策风险水平，并通过反馈机制降低内部决策风险水平。显着性检验用于确定标准权重并衡量子组的外部决策风险水平。然后将内部和外部决策风险级别合并以得出子组权重。置信区间用于将样本数据转换为区间数，然后对区间数进行汇总和分析以得出决策结果。同时，在整个决策过程中会通过各种方式考虑风险态度。通过案例分析和比较分析，以及敏感性分析，来说明该方法的可行性和合理性。我们的实验表明，决策风险和风险态度与大集团决策息息相关。