A fuzzy rough copula Bayesian network model for solving complex hospital service quality assessment

Area of Science:

• Healthcare Management
• Decision Science
• Artificial Intelligence

Background:

• Hospitals are central to healthcare, with service quality being a critical factor.
• Decision-making in healthcare is challenged by interdependencies, dynamic features, and uncertainties.
• Existing methods struggle to address both objective and subjective uncertainties in complex systems.

Purpose of the Study:

• To develop an advanced decision-making approach for assessing hospital service quality.
• To integrate Bayesian networks, copula, and fuzzy rough sets for robust analysis.
• To address dynamic features and both objective and subjective uncertainties in quality assessment.

Main Methods:

• Utilized a Bayesian copula network to model interrelationships and joint probability distributions.
• Employed fuzzy rough set theory with neighborhood operators to manage subjective decision-maker evidence.
• Developed a novel framework combining Copula Bayesian Network and extended fuzzy rough set techniques.

Main Results:

• The proposed method effectively reduces uncertainty in decision-making processes.
• Demonstrated the ability to assess dependencies between various factors in hospital service quality.
• Validated through a real-world analysis of hospital service quality in Iran.

Conclusions:

• The integrated approach offers a powerful tool for complex decision-making in healthcare.
• The method successfully handles dynamic features and subjective/objective uncertainties.
• Provides a novel framework for ranking alternatives based on multiple criteria in quality assessment.