论文部分内容阅读
针对比率标度在描述两两比较结果上的不足以及非结构化比较方式带来的不一致问题,提出了一种新的基于两两比较的多属性决策方法——认知最优最劣方法(CBWM).CBWM将最优和最劣的对象与其他对象分别进行比较,用差值标度度量比较结果,得到2个成对的比较向量,并基于最小化最大偏差和最小化总平方偏差2个模型对权重进行推断.CBWM需要的两两比较次数为2n-3,显著地少于P-CNP和AHP的n(n-1)/2,而且CBWM可以更方便地进行一致性检验,同时,一致性指数的对比分析显示CBGWM的一致性优于P-CNP;在职称评审中的应用结果显示CBWM得到的结果的总偏差显著地小于BWM,这表明CBWM可以更好地反映决策者对问题的认知,得到更加可靠的决策结果.理论分析和实践应用都证明了CBWM的有效性.
Aiming at the shortcomings of describing the comparison results and the inconsistencies caused by the unstructured comparison, a novel multiple attribute decision making method based on pairwise comparisons is proposed. CBWM) .CBWM compares the best and worst objects respectively with other objects, and uses the difference scale to measure the comparison results to get two pairs of comparison vectors, and based on the minimized maximum deviation and the minimized total square deviation 2 The models deduce the weights, CBWM requires 2n-3 comparisons, which is significantly less than n (n-1) / 2 for P-CNP and AHP, and CBWM makes it easier to perform consistency checks, , The consistency analysis of consistency index showed that the consistency of CBGWM was better than that of P-CNP. The application results of professional title evaluation showed that the total variance of CBWM was significantly smaller than that of BWM, which indicated that CBWM could better reflect the policy makers’ And get more reliable decision-making results.The theoretical analysis and practical application have proved the effectiveness of CBWM.