一种基于信度函数的综合评价方法

综合评价问题的本质是把被评价对象的物理可测指标转换为较抽象的功能性指标。其主要困难是这种转换或计算所依赖的假设和逻辑基础不易建立。以Dempster-Shafer证据理论为逻辑基础 ,建立了不确定性评价问题的一种置信度模型 ,给出了把下层指标评价建模为信度函数并应用Dempster规则合成为上层指标综合评价的方法

An Evaluation Method Based on Belief Functions

The essence of synthetic evaluation problems is to transform those physically measurable attributes of the object under evaluation into its functional attributes which are usually abstract and synthetic. The major difficulty lies in the fact that the assumptions and the logical foundation underlying the above transformation are not able to easily set up. By using Dempster-Shafter theory of evidence as the logical foundation, this paper constructs a confidence-degree model for uncertain evaluation problems, and presents an approach by which the index evaluations are modeled as belief functions and combined using Dempster rule to obtain the synthetic evaluation of the higher-layered indice.