In large group decision-making, participators with different knowledge structures, backgrounds, and other characteristics are unlikely to accurately evaluate alternatives. For this, it is necessary to decompose alternatives into several elements, and consider the participators’ preferences for elements of alternatives and the multiple classifications for participators according to their characteristics. However, related studies are still scarce. The objective of this paper is to propose a multi-elemental large group decision-making method, in which the desirable alternative(s) are selected from a set of feasible alternatives according to the preference information on multiple elements of alternatives provided by participators from multiple subgroups, and multiple classifications for participators are considered. In the method, according to the strict preference ordering of elements provided by participators, the percentage distributions on preferences of each subgroup concerning each element are firstly presented under each classification for participators. Secondly, the decision weight of each subgroup is determined by three factors, i.e., the consensus of preferences provided by each subgroup, the organizer’s preference for each subgroup, and the number of participators in each subgroup. Then, the comprehensive preference concerning each element is determined by combing the preference information from multiple subgroups and the decision weights of multiple subgroups, the overall preference vector can be obtained under each classification, and the virtual alternatives are determined by normalizing the overall preference vector. Further, considering multiple classifications for participators, the overall dominant degrees of alternatives can be obtained by calculating the similarity degrees between each virtual alternative and each alternative, thus the ranking order of alternatives can be obtained based on the overall dominant degrees of alternatives. Finally, an example is given to confirm the feasibility of the proposed method. The results of the sensitivity and comparative analyses show that the proposed method is applicable and effective. The proposed method can further enrich and improve the theory and approach of large group decision-making with multiple elements considering multiple classifications for participators.

在大型群体决策中，具有不同知识结构、背景和其他特征的参与者不太可能准确评估替代方案。为此，需要将备选方案分解为若干要素，并考虑参与者对备选方案要素的偏好以及根据参与者的特征对参与者进行多重分类。然而，相关研究仍然匮乏。本文的目的是提出一种多元素大群体决策方法，根据多个参与者提供的对多个备选方案元素的偏好信息，从一组可行备选方案中选择所需的备选方案。考虑了参与者的子组和多重分类。在该方法中，根据参与者提供的元素的严格偏好排序，首先在参与者的每个分类下呈现每个子组对每个元素的偏好的百分比分布。其次，每个子组的决策权重由三个因素决定，即每个子组提供的偏好共识、组织者对每个子组的偏好以及每个子组的参与者数量。然后，结合来自多个子组的偏好信息和多个子组的决策权重，确定每个元素的综合偏好，得到每个分类下的总体偏好向量，并通过对总体偏好向量进行归一化来确定虚拟选项。更远，考虑到参与者的多重分类，通过计算每个虚拟备选方案与每个备选方案之间的相似度，可以得到备选方案的总体主导度，从而根据备选方案的总体主导度得出备选方案的排名顺序。最后给出算例验证了所提方法的可行性。敏感性分析和对比分析结果表明所提方法的适用性和有效性。该方法可以进一步丰富和完善考虑参与者多重分类的多要素大群体决策理论和方法。通过计算每个虚拟备选方案与每个备选方案之间的相似度，可以得到备选方案的总体主导度，从而根据备选方案的总体主导度得出备选方案的排名顺序。最后给出算例验证了所提方法的可行性。敏感性分析和对比分析结果表明所提方法的适用性和有效性。该方法可以进一步丰富和完善考虑参与者多重分类的多要素大群体决策理论和方法。通过计算每个虚拟备选方案与每个备选方案之间的相似度，可以得到备选方案的总体主导度，从而根据备选方案的总体主导度得出备选方案的排名顺序。最后给出算例验证了所提方法的可行性。敏感性分析和对比分析结果表明所提方法的适用性和有效性。该方法可以进一步丰富和完善考虑参与者多重分类的多要素大群体决策理论和方法。敏感性分析和对比分析结果表明所提方法的适用性和有效性。该方法可以进一步丰富和完善考虑参与者多重分类的多要素大群体决策理论和方法。敏感性分析和对比分析结果表明所提方法的适用性和有效性。该方法可以进一步丰富和完善考虑参与者多重分类的多要素大群体决策理论和方法。