基于交叉效率DEA和随机模拟的区间语言偏好关系排序方法

针对区间加性语言偏好关系排序问题,提出一种基于交叉效率DEA和随机模拟的排序向量计算方法.首先,针对具有一致性的加性语言偏好关系,构建产出导向的DEA模型,分析其相对效率得分与排序向量之间的内在关联.其次,对不满足一致性的加性语言偏好关系,建立改进仁慈型语言偏好关系交叉效率DEA模型,并提出基于该模型的加性语言偏好关系的排序方法.最后,将构建的DEA模型与随机偏好分析相结合,提出区间加性语言偏好关系期望排序向量和可信度的Monte Carlo计算方法.算例分析表明,本文提出的方法能够有效避免信息损失,具有较好的可信度和适用性.

The ranking method of interval linguistic preference relations based on cross efficiency DEA and stochastic simulation

For the ranking problem of interval additive linguistic preference relation, we propose a new method to calculate the priority vector based on cross efficiency DEA and stochastic simulation. Firstly, the output-oriented DEA model is constructed for the additive consistent preference relations, and the intrinsic correlation between the relative efficiency score and the ranking vector is analyzed. Secondly, we develop a revised benevolent cross-evaluation DEA model for the inconsistent additive linguistic preference relation and a sorting method of additive linguistic preference relation is proposed based on this model. Finally, the proposed DEA model is combined with stochastic preference analysis, and a Monte Carlo calculation method for the expected priority vector and confidence of the interval additive linguistic preference relations is proposed. The numerical results show that the proposed method can effectively avoid the loss of information and has good reliability and applicability.