摘要
针对大型群组多属性决策问题,给出了备选对象的优势集和Pareto有效率,并讨论了二者的性质.证明并指出了只有备选对象为Pareto解时,其Pareto有效率才可能不为0.将Pareto备选对象的Pareto有效率作为其"最优决策"的先验概率分布,然后利用Bayes公式和群组专家们决策的后验概率对其加以修正,即可得到"最优决策"概率最大的备选对象.该方法在充分利用专家组决策信息的前提下避免了寻找一个主观集结规则的决策问题,不需要集结出一个权重结果,从而减少了决策过程中主观因素的影响,并且当将每位专家的决策看成一个独立的随机实验时,理论上专家人数越多,决策结果越精确.最后以一个算例说明了所提出方法的有效性.
Aiming at the multi-attribute large group decision-making,this paper presents the superior set of candidates and Pareto valid probability and then discusses the characteristics of them.The results show that the candidate set may not be zero only if it is the Pareto solution.This paper takes candidate’s Pareto valid probability as the priori probability distribution of ＂optimal decision＂ and then corrects them using the posterior probability in Bayes formula and experts’ decisions in groups.As a result,the candidate with the largest probability of ＂optimal decision＂ can be achieved.This method avoids finding a subjective aggregation rules after taking full advantage of the information from experts in group decision-making.More experts in groups can make the more accurate result when reducing subjective factors in decision-making process and taking each expert’s decision as an independent random experiment.Finally,a numerical example shows the effectiveness of the proposed method.
出处
《控制与决策》
EI
CSCD
北大核心
2013年第7期1051-1054,共4页
Control and Decision
基金
国家自然科学基金创新群体项目(70921001)
国家自然科学基金面上项目(71072078
70971139)
国家自然科学基金青年项目(71103203)
教育部人文社会科学基金项目(09YJC790260)
中南大学人文社科杰出青年人才基金项目(2011RWSK008)
中南大学2011年青年教师助推专项基金项目(2011QNZT237)
