A PSO-algorithm-based consensus model with the application to large-scale group decision-making

Group decision-making (GDM) implies a process of extracting wisdom from a group of experts. In this study, a novel GDM model is proposed by applying the particle swarm optimization (PSO) algorithm to simulate the consensus process within a group of experts. It is assumed that the initial positions of decision-makers (DMs) are characterized by pairwise comparison matrices (PCMs). The minimum and maximum of the entries in the same locations of individual PCMs are supposed to be the constraints of DMs’ opinions. The novelty comes with the construction of the optimization problem by considering the group consensus and the consistency degree of the collective PCM. The former is to minimize the distance between the collective PCM and each individual one. The latter is to make the collective PCM be acceptably consistent in virtue of the geometric consistency index. The fitness function used in the PSO algorithm is the linear combination of the two objectives. The proposed model is applied to solve a large-scale GDM problem arising in emergency management. Some comparisons with the existing methods reveal that the developed model has the advantages to decrease the order of an optimization problem and reach a fast yet effective solution.