Risk-Averse Two-Stage Stochastic Minimum Cost Consensus Models with Asymmetric Adjustment Cost

Area of Science:

• Operations Research
• Decision Sciences
• Computational Economics

Background:

• Group decision-making involves coordinating diverse views to reach consensus.
• Uncertain factors significantly influence consensus building, introducing risks and potential losses.
• Existing models often fail to adequately address both uncertainty and risk in consensus processes.

Purpose of the Study:

• To develop novel two-stage mean-risk stochastic minimum cost consensus models (MCCMs) incorporating asymmetric adjustment costs.
• To investigate three distinct modeling methods for these MCCMs.
• To address the limitations of current models in handling uncertainty and risk during consensus formation.

Main Methods:

• Development and application of three modeling methods for two-stage mean-risk stochastic MCCMs with asymmetric adjustment costs.
• Utilization of the L-shaped algorithm for solving complex optimization models.
• Validation through a numerical example on a peer-to-peer online lending platform and comparison with CPLEX solver.

Main Results:

• The L-shaped algorithm demonstrates accuracy and efficiency in solving the proposed MCCMs, verified against the CPLEX solver.
• Sensitivity analyses confirm the impact of risk on consensus model outcomes.
• Comparisons highlight the advantages of the new risk-averse MCCMs over existing stochastic and robust consensus models under asymmetric costs.

Conclusions:

• The proposed modeling approach effectively integrates uncertainty and asymmetric costs into minimum cost consensus models.
• The L-shaped algorithm provides a reliable and efficient method for optimizing these complex consensus-building scenarios.
• The findings offer valuable insights for improving group decision-making processes in uncertain environments, particularly in financial platforms.