基于两阶段的物流系统综合评价DEA/AHP法

物流系统具有多个属性指标 ,对物流系统进行综合评价 ,往往要求事先设定这些属性的权重。为了减少人为因素的主观影响 ,提高评价的客观性 ,在构建物流系统评价指标体系的基础上 ,将DEA和 AHP算法融合 ,建立两阶段评估模型。第一阶段计算两两实体构成的输入输出系统有效系数 ;第二阶段对这些有效系数构成的 AHP判断矩阵 ,计算各个实体的全排序值。这种综合评价方法充分利用了 AHP方法和 DEA方法的优点 ,避免了人为设定 AHP判断矩阵的主观性和随意性 ,DEA不能在一个尺度上全排序的缺点 ,最后设计了一个算例 ,验证了评价方法的可行性和一致性。

Logistics synthesis evaluating on DEA/AHP two-stage model

Performance evaluation of a group of entities is frequently based on the values of several attributes and usually requires the weights of the attributes to be set in advance. After an index of logistics system being built and with the Data Envelopment Analysis(DEA) algorithm and the Analytical Hierarchical Process(AHP) algorithm being integrated, a two stage hybrid evaluating model is suggested in which no weight needs to be assigned factitiously.In the first stage of the model the DEA is used on each pair of units to create a pair wise evaluation matrix while in the second stage the AHP is applied to process the evaluation matrix and then the full units ranking is obtained. The hybrid model takes the best advantages of both AHP and DEA and at the same time,avoids either the subjectivity of AHP or the dichotomy of DEA. Finally, a practical case is also computed to validate the consistency of the hybrid evaluation algorithm.