论文部分内容阅读
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.
For the large group multi-attribute decision making problem, the dominant set and Pareto efficient of the candidate are given and the properties of the two are discussed. It is proved that the Pareto efficient is only possible when the candidate is a Pareto solution Is not 0. The Pareto efficiency of the Pareto candidate is taken as the prior probability distribution of its “optimal decision”, and then corrected by the Bayesian formula and the posterior probability of the group expert decision-making, we can get “Optimal decision ” which has the largest probability of probability, which avoids the decision-making problem of finding a subjective agglomeration rule while making full use of the expert decision-making information, and does not need to build a weighted result, thus reducing the decision-making process Subjective factors, and when each expert’s decision-making is regarded as an independent randomized experiment, the more theoretically experts, the more accurate decision-making result.Finally, an example is given to show the effectiveness of the proposed method.