不完全信息下多属性决策的证据推理方法

his paper studies a kind of multi-attribute decision making problems with unknown weights and missing values. A method based on evidential reasoning for the mentioned multi-attribute decision making problems is proposed. Firstly, it gets the basic probability assignment by standardizing the decision matrix values, confirming focal element under different attribute and computing their score values. Then, by employing evidential combination defined in Dempster-Shafer theory, the ranking of the whole alternatives can be attained through trust interval computing and ranking. At last an example is given to explain the feasibility and availability of this method.