基于信息云组合权重与灰色关联度改进的TOPSIS

Improved TOPSIS Based on Combination Weight of Information Cloud and Grey Ideal Cloud

针对多属性决策问题,考虑指标权重的不确定性与欧式距离的缺陷,提出了一种基于虚拟最劣解、灰色关联度以及信息云组合权重改进的TOPSIS方法。该方法以TOPSIS模型为基础,首先由逆向云发生器确定评价指标的权重云;其次,通过虚拟最劣解与灰色关联理论完成正、负理想灰色加权关联系数矩阵的构建,继而获得待选方案的综合云;最后,运用兼顾形状与距离的相似度测算方法,获得待选方案与理想解之间的综合云相似度,并以“相似程度”完成“距离”的替换,提出了一种新的贴近度计算方法,实现了方案的比选。实例结果表明:与其他决策模型相比,该方法既能考虑待选方案与理想解之间的位置关系与态势变化,还能克服排序过程的不合理现象,决策结果合理有效。

In this paper, we propose an improved TOPSIS method based on virtual worst solution, grey correlation degree and information cloud combination weight method for the purpose of overcoming the uncertainty of index weight and the defect of euclidean distance. This method is based on TOPSIS, the weight cloud of evaluation index is determined by reverse cloud generator. Then, the construction of positive and negative ideal grey correlation coefficient cloud matrix is completed by virtual worst solution and grey correlation coefficient, and the comprehensive cloud of comparison scheme is obtained. Finally, the similarity calculation method considering shape and distance is used to calculate the similarity between the comprehensive cloud of the candidate scheme and the ideal scheme, a new closeness algorithm is constructed by using similarity instead of distance, and next the scheme comparison is realized. The example results show that compared with other decision models, this method can not only consider the position relationship and situation change between the candidate scheme and the ideal solution, but also overcome the unreasonable phenomenon in the sorting process, and the decision results are accurate and effective.