Large group decision-making (LGDM) with the participators from multiple subgroups of stakeholders: A method considering both the collective evaluation and the fairness of the alternative

Large group decision-making (LGDM) with the participators from multiple subgroups of stakeholders is a valuable research topic with many practical backgrounds. In the LGDM analysis, not only the collective evaluation but also the fairness of the alternative among the stakeholders from different subgroups needs to be considered, whereas the fairness of the alternative is neglected in the existing studies. This paper proposes a method for LGDM considering both the collective evaluation and the fairness of the alternative. In the method, according to the Pauta criterion (3 sigma criterion), the noise of evaluations provided by participators from each subgroup is reduced. Then, a cumulative distribution on collective evaluation of each alternative is determined by aggregating the cumulative distributions on evaluations and decision weights of multiple subgroups. The ranking value of each alternative considering the collective evaluation is measured by calculating the distances of the cumulative distribution on collective evaluation from the ideal and anti-ideal cumulative distributions, simultaneously. Further, the fairness degree of an alternative between two subgroups is defined according to the distance between the cumulative distributions on evaluations provided by the participators from the two subgroups. The ranking value of each alternative considering the fairness is measured by aggregating the fairness degrees of the alternative between any two subgroups. According to the ranking values considering the collective evaluation and the ranking values considering the fairness, the ideal and anti-ideal alternatives are defined. The ranking value of each alternative considering both the collective evaluation and the fairness is determined according to the closeness coefficient of each alternative. Finally, an example is given to illustrate the use of the proposed method.