不完全权重信息下多属性决策方法研究

针对目标属性的权重不完全确知的情况下，提出了一种先进行局部优化，再组合赋权并对方案进行排序的决策方法。此方法是从最小偏差的角度出发，构造了目标权重的统计量，它既利用了子样的信息，又利用了先验信息，评价结果客观可靠。另外，在此基础上进一步加入权系数的信息熵，建立了优化的数学模型。最后，通过实例表明新方法的有效性。

Study of Multi-attribute Decision - Making in the Case ofPartial Weight Information

This paper presents a method for MADM of partial weight information. The method consists of three steps --- first optimizing regionally, then forming assigned weigh and ranking all decisions. The method, according to the minimum deviation, constructs the statistics - variable of weight and contains the utilization of both samples information and previous information so as to make the evaluations objective and believable. In addition, IE is employed and optimal mathematics model is set up on the basis of the method. At last, an illustrative example is given to demonstrate the effectiveness of the new method.