Optimizing critical care capacity using fuzzy multi-objective digital twins and deep learning

Capacity management in intensive care units is one of the most complex challenges in the health system, requiring advanced decision-making approaches due to uncertainty in demand, resource constraints, and clinical sensitivity of decisions. This study presents an innovative framework for optimizing intensive care capacity, in which data-driven predictions, fuzzy modeling, and multi-objective decision-making are integrated within a dynamic digital twin. The proposed framework supports adaptive and uncertainty-aware decision-making by utilizing patient behavior prediction and continuous system status updates. Numerical results based on real and simulated data show that the proposed framework reduces the clinical risk index, admission delay, and workload imbalance, while providing a more controlled use of ICU capacity and faster recovery after demand shocks, compared to the fixed rule-based policy and the deterministic optimization model. The findings of this study show that integrating intelligent decision-making with digital twins can play an effective role in the development of decision support systems in the field of intensive care.