Consensus of large-scale group decision making in social network: the minimum cost model based on robust optimization

Area of Science:

• Decision Science
• Network Science
• Optimization

Background:

• Large-scale group decision making (LSGDM) is emerging in social networks.
• Uncertainty in expert adjustment costs poses challenges in practical LSGDM consensus.
• Existing models often overlook the social relationships among experts.

Purpose of the Study:

• To develop a novel consensus model for LSGDM in social networks using robust optimization.
• To address the challenge of uncertain unit adjustment costs for experts.
• To incorporate social network structures and expert relationships into the decision-making process.

Main Methods:

• Expert clustering based on trust degree and relationship strength to form subgroups.
• Development of a consensus index to quantify the harmony and consensus level among experts.
• Formulation of a minimum cost model utilizing robust optimization to solve consensus problems.

Main Results:

• The proposed model effectively handles uncertain expert costs in LSGDM.
• Expert clustering based on social relationships improves consensus efficiency.
• The consensus index provides a reliable measure of group agreement.
• Case studies demonstrate the validity and advantages of the robust optimization approach.

Conclusions:

• The robust optimization consensus model offers a viable solution for LSGDM with uncertain costs.
• Integrating social network analysis enhances the consensus-building process.
• The method provides a practical framework for achieving consensus in complex group decisions.