Fairness-aware large-scale collective opinion generation paradigm: A case study of evaluating blockchain adoption barriers in medical supply chain

Abstract

Generating collective opinion based on probability distribution function aggregation occupies crucial roles in accomplishing probabilistic risk analysis, probabilistic-forecast-based prediction, and uncertain assessment tasks. However, rare efforts have been paid to exploring collective opinion generation based on massive quantified judgments from large-scale experts. In this study, we establish a novel large-scale collective opinion generation paradigm based on probability distribution function aggregation to accomplish complicated assessment and evaluation tasks in decision analysis. To this end, we reformulate the existing bi-objective optimization model with the exclusion of the consensus dimension and inclusion of fairness concern among subject matter experts. The formulation of collective fairness utility uses the notion of aggregation functions and contributes to the establishment of large-scale collective opinion generation paradigm with capabilities of modeling distinct fairness distributions among the subject matter experts. The established fairness-aware large-scale collective opinion generation model advocates an adjusted bi-objective optimization model that maximizes the confidence level and fairness utility within the decision group for opinion aggregation on both intra-group and inter-group levels. It is endowed with the flexibility to be combined with context-specific expert classification and criterion weight assignment techniques to generate powerful large-scale group decision making prototypes that accommodate with diversifying decision-making scenarios. We apply the proposed fairness-aware large-scale collective opinion generation framework to the assessment of blockchain adoption barriers in medical supply chain to demonstrate its rationality and effectiveness.