Modeling a data-driven robust maximum expert consensus framework with fairness concern under uncertain environments

Wu, Linfan, Qu, Shaojian, Luan, Dongqing, Wahab, M. I. M., Ji, Ying
Operational Research · Springer Nature · 2026-09-24 · DOI: 10.1007/s12351-026-01110-8

Abstract

The consensus optimization model serves as an effective approach for addressing group decision-making (GDM) problems. In the consensus reaching process (CRP), linear cost functions fail to adequately capture the dynamic nature of individual adjustment costs. Additionally, equity theory suggests that individuals consider not only their own benefits but also those of others, making relative comparisons. Moreover, fluctuations in the decision-making environment have introduced uncertainty in unit adjustment costs. These challenges complicate the development of consensus optimization models. To address the above issues, we introduce the quadratic cost into the maximum expert consensus model (MECM) and add the fairness utility level constraint to ensure that the model outcomes satisfy predefined equity criteria, reflect fairness-sensitive behavior among individuals, and better align with real-world decision-making contexts. In addition, we employ a robust optimization (RO) approach to address the intrinsic uncertainty of unit adjustment costs and, based on this, leverage historical data by implementing a data-driven RO approach to further address uncertainty. Finally, the proposed model is applied to an empirical case involving urban and rural compulsory education subsidies to validate its practical effectiveness. The superiority of the proposed model is substantiated through comparative and sensitivity analyses.

Read at publisher →

Similar articles