Abstract

Efficient hospital bed management is critical for optimizing healthcare resources and reducing patient wait times; however, traditional bed allocation methods, which assign beds to specific specialties, often lead to inefficiencies due to rigid compartmentalization and lack of real-time visibility. This study explores the application of Digital Twin (DT) technology as an innovative solution that enables dynamic, fully pooled bed allocation across specialties, enhancing flexibility and resource utilization. Using three years of admission data (2021–2023) from a large hospital in the GCC, two optimization models are compared: a traditional specialty-based allocation and a DT-enabled dynamic pooling system. A simulation-based optimization approach, combining Discrete-Event Simulation, Genetic Algorithm, and Monte Carlo analysis, evaluates both models against key performance indicators, including total beds required, patient waiting times, and bed utilization efficiency. The DT-enabled strategy demonstrated remarkable improvements, reducing total bed requirements by 53.7% (from 201 to 93 beds), decreasing patient waiting times by 63.9% (from 24.38 to 8.80 days), and significantly enhancing bed utilization efficiency through more balanced resource distribution. These results highlight substantial cost savings and operational gains, though implementing DT-based pooling requires overcoming challenges related to staff training, workflow adjustments, and cultural shifts within hospitals. A gradual, phased implementation is recommended to ease the transition from traditional specialty-based models to pooled systems. Furthermore, DT technology holds potential beyond bed management, offering opportunities in predictive analytics, capacity planning, and inter-hospital collaboration. This study provides robust evidence for adopting DT-enabled bed pooling to improve hospital efficiency and patient care outcomes.