Abstract: In view of the complexity and multi-period characteristic of real decision-making problems, the traditional multi-attribute decision-making method is insufficient to solve intricate real decision-making problems. Dynamic multi-attribute group decision-making method as a powerful tool to address this type of decision-making problems is proposed, which involves multiple periods and multiple experts. In the decision-making process, there are two important aspects that influence the final decision results: the weighting information and the bounded rationality of decision makers.
With respect to interactive dynamic multi-attribute group decision-making problems, where the weighting information is incompletely known, and intuitionistic fuzzy numbers express attribute values, this paper proposes a hybrid decision-making method: prospect theory+VIKOR method. Considering the interactive characteristic of the dynamic multi-attribute group decision-making problem, a generalized Shapley intuitionistic fuzzy Frank-Choquet average operator is introduced to infuse interactive evaluation information. Additionally, prospect intuitionistic fuzzy decision-making matrices are calculated by prospect theory to cope with the bounded rationality of decision-makers. Furthermore, how to obtain the decision weighting information is also studied. The corresponding optimization models are constructed using the intuitionistic fuzzy cosine similarity measure and the Shapley function to get the optimal weights of experts, periods, and attributes. According to these results, an interactive dynamic intuitionistic fuzzy multi-attribute group decision-making method is proposed.
The new method is applied to a case study: evaluating the high-level treatment of general industrial solid waste in zero-waste city construction. The decision results show that the new method has some merits for addressing complex decision-making problems by comparing the results with the existing methods. It indicates the importance of considering various weights’ interactions and decision makers’ bounded rationality.
To ensure the final results to represent the decision makers’ opinions with the preset level, we can further incorporate the consensus into the new method. In addition, we can extend the new method to more complex setting, including large-scale group decision-making and group decision-making under social networks.

Key words: dynamic multi-attribute group decision-making, intuitionistic fuzzy variable, prospect theory, VIKOR method, zero-waste city

摘要： 针对具有交互作用权重信息不完全已知,且评价值以直觉模糊数形式给出的动态多属性群决策问题,本文提出一种混合决策方法:前景理论+VIKOR法。首先,基于前景理论提出一种直觉模糊加权决策矩阵构造方法。其次,为求解未知权重信息,基于直觉模糊余弦相似性测度构建规划模型。同时,为对具有交互作用的评价信息进行融合,提出广义Shapley直觉模糊Frank-Choquet累积算子。最后,将新方法运用到“无废城市”建设中。

关键词: 动态多属性群决策, 直觉模糊变量, 前景理论, VIKOR法, “无废城市”