Real-Time Hospital Bed and Emergency Supply Demand Sensing Using Crisis-Context Aspect Sentiment and Spatiotemporal Edge Neural Networks

• Billy Elly

  Lautech

  Author

Hospital resource management faces critical challenges during public health emergencies, with traditional forecasting methods failing to capture the non-linear, spatiotemporal dynamics of demand surges and supply chain disruptions . Existing approaches primarily rely on static inventory policies and historical averages, which prove inadequate when crisis-driven demand patterns deviate significantly from historical norms . This study addresses this gap by proposing a novel hybrid framework that integrates crisis-context aspect sentiment analysis with spatiotemporal edge neural networks for real-time hospital bed and emergency supply demand sensing. The methodology employs a collaborative CNN-LSTM architecture deployed across edge-cloud infrastructure, achieving 92.3% prediction accuracy for bed occupancy forecasting and 89.7% accuracy for emergency supply demand prediction across four major US hospital networks . The framework demonstrates a 34% improvement in lead time for critical supply alerts compared to conventional methods and reduces prediction latency by 41% through edge computing optimization. These findings establish a replicable, scalable framework for hospital resource management with significant implications for healthcare administrators, emergency preparedness policymakers, and the broader field of healthcare operations research

• pdf

Copyright (c) 2026 Billy Elly (Author)

This work is licensed under a Creative Commons Attribution 4.0 International License.