An intuitionistic fuzzy automated negotiation model for personalized and efficient shared decision-making

Area of Science:

• Health Informatics
• Artificial Intelligence in Healthcare
• Decision Science

Background:

• Shared decision-making (SDM) integrates patient preferences with medical expertise to enhance patient involvement and satisfaction.
• Current SDM tools struggle with complex medical information, ambiguous patient preferences, and uncertain outcomes, leading to lengthy negotiations.
• Existing models often fail to simultaneously address these multifaceted challenges in healthcare decision-making.

Purpose of the Study:

• To introduce a novel Agent-based Auto-negotiation Model based on Intuitionistic Fuzzy Sets (AN-IFF) to overcome SDM challenges.
• To enhance the efficiency and effectiveness of the negotiation process in shared decision-making.
• To model diverse decision-making behaviors and improve outcomes in patient-provider interactions.

Main Methods:

• Developed an Agent-based Auto-negotiation Model (AN-IFF) utilizing intuitionistic fuzzy sets and fuzzy inference systems to handle uncertainty.
• Incorporated a time-discounting mechanism for dynamic concession strategy adjustment and optimal counter-offer generation.
• Integrated three distinct negotiation strategies to simulate optimistic, balanced, and pessimistic decision-maker personalities within the fuzzy modeling framework.

Main Results:

• AN-IFF effectively models personality-driven differences in concession behavior within shared decision-making contexts.
• The model demonstrated significant improvements in joint satisfaction and fairness compared to baseline methods.
• AN-IFF reduced the number of negotiation rounds, indicating increased efficiency in the decision-making process.

Conclusions:

• The AN-IFF model offers a robust solution for managing uncertainties and diverse behaviors in shared decision-making.
• This approach enhances negotiation outcomes, leading to greater patient satisfaction and fairer agreements.
• AN-IFF represents a significant advancement in utilizing AI for optimizing healthcare decision-making processes.