基于Jaynes最大熵的模糊多属性群决策投影方法

发布时间：2019-04-22 18:51

【摘要】：为了解决属性权重完全未知的模糊多属性群决策问题,采用基于Jaynes最大熵原理的投影偏差模型确定指标权重,运用基于多准则妥协解排序法的妥协解排序框架实现决策过程。首先运用区间直觉模糊集表达决策者的犹豫模糊评价信息,然后运用集成算子获得综合决策矩阵,再运用Lingo11~汶求解获得属性权重,利用信息集结获得备选方案的加权投影,最后根据方案的最大群效应值、最小后悔值和综合指标值进行方案排序。实验显示,方案的排序结果与其它方法保持一致,趋势图显示此方法计算的各方案综合指标值变化趋势更明显。分析表明,该方法避免了其他权重确定方法的不足之处,消除了计算过程中的折衷损失,适用于属性权重完全未知的多属性群决策方案排序问题的评价。
[Abstract]:In order to solve the fuzzy multi-attribute group decision-making problem with unknown attribute weights, the projection deviation model based on the Jaynes maximum entropy principle is used to determine the index weights, and the compromise solution ordering framework based on the multi-criteria compromise solution ordering method is used to realize the decision-making process. Firstly, the interval intuitionistic fuzzy set is used to express the decision maker's hesitant fuzzy evaluation information, then the synthetic decision matrix is obtained by using the integration operator, then the attribute weight is obtained by using the Lingo11~ Wen solution, and the weighted projection of the alternative scheme is obtained by using the information aggregation. Finally, the scheme is sorted according to the maximum group effect value, the minimum repentance value and the synthetic index value of the scheme. The experimental results show that the sorting results of the scheme are consistent with other methods, and the trend chart shows that the change trend of the synthetic index values of each scheme calculated by this method is more obvious. The analysis shows that the proposed method avoids the shortcomings of other weight determination methods and eliminates the compromise loss in the calculation process. It is suitable for evaluating the scheduling problem of multi-attribute group decision-making with completely unknown attribute weights.
【作者单位】：中国矿业大学矿业工程学院;
【基金】：国家自然科学基金资助项目(51505480,71401164) 江苏省自然科学基金资助项目(BK20150197)
【分类号】：C934

【相似文献】