AI-DRIVEN PREDICTIVE OPERATIONS MANAGEMENT: A BUSINESS SCIENCE FRAMEWORK FOR DYNAMIC HOSPITAL RESOURCE OPTIMIZATION AND CLINICAL WORKFLOW EFFICIENCY

Objective: This study aims to develop and evaluate an AI-driven, business-science-based predictive operations management framework to optimize hospital resource allocation and enhance clinical workflow efficiency, particularly under dynamic and high-demand conditions. Theoretical Framework: Grounded in the integration of healthcare informatics, operations research, and machine learning, the proposed framework applies predictive analytics and optimization to transform hospital operations from reactive to proactive management. Method: The framework was implemented in a simulated hospital environment using real and synthetic datasets, including MIMIC-IV and World Health Organization (WHO) hospital resource utilization data. Machine learning models (XGBoost, LSTM) were employed for patient demand forecasting, supported by optimization algorithms and Monte Carlo simulations to validate system robustness under varying load conditions. Results and Discussion: The implementation improved bed utilization from 73% to 88%, reduced staff idle time by 15%, and decreased surgery scheduling conflicts by 23%. K-means clustering identified operational bottlenecks, while Monte Carlo simulations confirmed high system stability and adaptability. These results demonstrate the framework’s capability to enhance operational efficiency and responsiveness. Research Implications: The findings provide evidence for integrating AI-driven predictive management systems in healthcare, offering administrators actionable insights for dynamic resource allocation and strategic planning. The approach is adaptable to single or multi-hospital networks and can support resilience during crises such as pandemics. Originality/Value: This research presents one of the first holistic, AI-enabled frameworks that combines real-time forecasting, optimization, bottleneck analysis, and simulation for end-to-end hospital operations management. The model’s scalability, adaptability, and validation across multiple performance indicators underscore its potential as a transformative tool for modern healthcare systems.