基于理想点的改进AHP综合排序算法

AHP是常用的多目标决策和评价方法,但是文献发现当被评价方案增加或减少时,层次分析法（AHP）不能保持其原有方案的排序结果,因此至今AHP法仍被广大学者所争议.本文研究了AHP中导致排序结果不一致现象的原因,发现当方案数变化时AHP在计算综合排序时,使用的计算数值会导致前后不一致,进而导致排序结果不一致现象的产生.本文提出了一种基于理想点贴近度的AHP综合排序算法可以很好地解决这一问题.AHP综合排序的典型算例证明了新算法的合理性.

Improvement on Algorithm for Final Priorities of AHP Based on Ideal Solution

Analytic hierarchy process(AHP) has been considerably criticized for possible rank reversal phenomenon caused by the addition or deletion of an alternative.This paper looks into the cause of rank reversal phenomenon and finds that rank reversal is caused by change of local priorities before and after an alternative is added or deleted.An approach is therefore proposed to keep the finall priorities unchanged to avoid rank reversal phenomenon.This approach use the TOPSIS closness as the final priorities index.At last a numerical examples are re-examined using the proposed approach to demonstrate its validity and practicability in rank preservation.