具有方案偏好权重完全未知的区间多属性决策方法

Interval multi-attribute decision-making method with scheme preference and completely unknown weight

针对属性值为区间数、权重信息完全未知且对方案的偏好信息为区间数的多属性决策问题,提出了基于相对熵与灰色关联分析的决策方法。首先,通过构建并求解一个单目标优化模型,得到属性权重值;然后,计算出每个方案客观偏好值与主观偏好值的相对熵和灰色关联系数,进而得到每个方案客观偏好与主观偏好的合成关联度,根据合成关联度对各方案进行排序优选;最后,通过房产投资决策的案例验证了该方法的可行性和有效性。

A multi-attribute decision making method based on relative entropy and grey relational analysis is proposed for multi-attribute decision making problems where the attribute value is interval number,the weight information is completely unknown and the preference information of the scheme is interval number.Firstly,the method obtains the attribute weight value by constructing and solving a single-objective optimiza⁃tion model;secondly,the relative entropy and gray correlation coefficient of objective preference value and sub⁃jective preference value of each scheme are calculated,and then the synthetic correlation degree between objec⁃tive preference and subjective preference of each scheme is obtained,according to the synthetic correlation de⁃gree,the schemes are sorted and selected;Fianly,the feasibility and effectiveness of the method are verified by the case of real estate investment decision.