针对具有多个属性集和信息类型的多属性群决策问题,运用模块化的思想,提出了一种新的决策方法。首先,将具有不同属性集与信息类型的决策矩阵视为相互独立的模块,按属性将其划分为若干子模块,并对子模块单独进行标准化；然后,计算每个模块中各方案的群体效用值、个体遗憾值和综合评价值。在此基础上,根据专家权重集结所得结果,并据此确定最优折衷方案。最后,通过两个算例验证了所提方法的可行性与有效性。

According to the multi-attribute group decision making problem with different attribute sets and information types, a new decision method is proposed. At first, decision matrixes with different attribute sets and information types are regarded as independent modules which are divided into many sub-modules according to attributes. All sub-module are standardized respectively according their information type. Then, the group utility value, the individual regret value and the comprehensive evaluation value of each alternative in each module are computed. Based on this, the results obtained above are aggregated and the optimal compromise alternative are determined. Finally, two numerical examples are presented to demonstrate the feasibility and effectiveness of the proposed method.

C934

国家自然科学基金重大研究计划重点支持项目 “非常规突发事件演化分析和应对决策的支持模型集成原理与方法” (编号：91024029 )。