Analyzing Consensus Measures in Group Decision Making

Abstract

In Group Decision Making (GDM) problems before to obtain a solution a high level of consensus among experts is required. Consensus measures are usually built using similarity functions measuring how close experts' opinions or preferences are. Similarity functions are defined based on the use of a metric describing the distance between experts' opinions or preferences. In the literature, different distance functions have been proposed to implement consensus measures. This paper presents analyzes the effect of the application of some different distance functions for measuring consensus in GDM. By using the nonparametric Wilcoxon matched-pairs signed-ranks test, it is concluded that different distance functions can produce significantly different results. Moreover, it is also shown that their application also has a significant effect on the speed of achieving consensus. Finally, these results are analysed and used to derive decision support rules, based on a convergent criterion, that can be used to control the convergence speed of the consensus process using the compared distance functions.

Attitude quantifier based possibility distribution generation method for hesitant fuzzy linguistic group decision making

“…(3) Consensus reaching approach is an important issue in group decision making problems. Many scholars have developed a number of methods that include: consensus measure of various types of preference [6,7,42]; feedback mechanism [30]; optimization-based consensus model [45,46]; consensus model with preference information [49,50]; dynamic consensus process model [21] and consensus model for large scale group decision making [15,18]. Our proposed method incorporates the attitude character towards reaching consensus for the first time and it supports experts adjust their evaluation quickly and automatically as it is taken care of by the systems, which reduces the direct number of adjustments in consensus reaching process required from experts.…”

“…In such cases, an effective consensus model would be necessary to bring closer the experts' preferences in an attempt to derive an acceptable group decision. Many scholars have developed a large number of consensus methods from different perspectives: consensus measure for various types of preferences [6,7,42]; feedback mechanism [30]; optimization-based consensus model [45,46]; consensus model with preference information [49,50]; dynamic consensus process model [21] and consensus model for large scale group decision making [15,18]. Herrera-Viedma et al in [12] presented a clear overview of recent achievements concerning 'soft consensus models' in a fuzzy environment, while Zhang et al carried out a comparative study on different consensus reaching approaches and proposed comparison criteria to evaluate the efficiency of consensus reaching processes [48].…”

Rogue Entrepreneurship

“…While a ''consensus'' may be (and has been) empirically identifiable (e.g., Ahlim et al, 2022;Wang et al, 2021), we simply note that, as with many fuzzy concepts, scholars (and entrepreneurs and their stakeholders) will choose cutoffs and measures suited to their research questions for methodological purposes. We also note that assessments of consensus can refer either to a degree of agreement (determining if consensus exists), as we do above, or it can refer to the distance of an opinion from the consensus (e.g., Chiclana et al, 2015), which correspond to our notion of ''how rogue'' an opinion or claim is. In the latter case, consensus measures are based on distance/similarity calculations, such as Euclidian (Chiclana et al, 2007), Cosine (Deza & Deza, 2009), and Jaccard (Salton & McGill, 1983) distance/similarity functions.…”

Similarity–trust network for clustering‐based consensus group decision‐making model

“…From this information, the closeness of experts' preferences can be assessed using similarities or distance functions. 2 González-Arteaga et al 3 proposed converting the preference relations into vectors to compute the correlation experts' preferences while Zhang et al 4 introduced the comparative linguistic expression preference relations to represent uncertain opinions of experts in GDM. The similarity measure of experts' preferences has been frequently used to evaluate the consensus level in the consensus process.…”

“…In general, these novel knowledge contributions are focused on the integration and impact of trust relation in SNGDM framework. Specifically, we (1) introduce a new combination function, where trust relation is integrated into the preference similarity network, (2) propose a new similarity-trust network to visualize the closeness of preferences and trust relationship between experts, (3) measure the centrality index in the similarity-trust network to identify the most important expert in the network, and (4) formulate a particular IOWA, similarity-trust centrality (STC)-IOWA, operator to aggregate all individual experts' preferences into a collective one according to the most important and trusted expert.…”