一类不确定信息下的多属性决策问题的研究  

Study on Multi-Attribute Decision-Making Problem with Uncertain Information

作　　者：廖貅武[1] 李垣[1] 董广茂[1] 

机构地区：[1]西安交通大学管理学院,西安710049

出　　处：《西安交通大学学报》2005年第8期792-795,803,共5页Journal of Xi'an Jiaotong University

基　　金：国家自然科学基金资助项目(70072022);中国博士后基金资助项目(2003(33)).

摘　　要：对一类不确定信息下的多属性决策问题进行了研究,给出了不确定信息的定义,即在决策过程中决策者不仅不能准确地估计方案的属性值,甚至不能确定同一属性下方案之间的偏好关系.在此基础上,提出使用具有置信结构的偏序关系来表示不确定信息,并给出了解决这类问题的模型和方法.该方法首先采用证据推理算法对不确定信息进行集成,然后使用优势和劣势这2种指数来确定方案间的偏好关系,并在这2种指数的定义中使用了偏序之间的距离公式.最后通过一个例子说明,在信息不确定的情况下该方法能够帮助决策者做出比较合理的决策.A class of multiple attribute decision-making problems with uncertain information was studied.The definition of the uncertain information was given, with which the decision-maker could neither estimate attribute values of alternatives precisely, nor determine the preference relations of alternatives under the same attribute. Based on above definition, the uncertain information was represented by partial ordering relations with belief structure. A new decision-making approach was proposed for solving such problems, in which an evidential reasoning algorithm was adopted to aggregate the uncertain information. The superior index and inferior index were used to establish the preference relation between alternatives. These two indexes were derived from the distance axioms of partial order. Finally, an example was given to demonstrate that the proposed approach can help decision-maker make reasonable decision under uncertain information.

关 键 词：多属性决策 不确定信息 证据推理 

分 类 号：TP11[自动化与计算机技术—控制理论与控制工程] C934[自动化与计算机技术—控制科学与工程]