International Journal of Science, Engineering and TechnologyAuthors: Aayush Chaudhary, Sumit Chakraborty, Nishant Sah, Abhiraj Chaudhary, Chinmay Dev, Mr. Bhupendra Ram
Abstract: Modern hospitals operate as complex adaptive systems where patient arrivals, bed occupancy, staff availability, and departmental load interact through tightly coupled feedback loops. Traditional hospital information systems provide fragmented, retrospective views of individual operational metrics, offering limited support for proactive capacity planning or what-if scenario analysis. This paper presents MedTwinX, a full-stack digital twin platform integrating real-time operational monitoring, discrete-event simulation (DES), statistical predictive analytics, heuristic reinforcement-learning-inspired optimization, role-based access control (RBAC), and AI model governance into a unified web application. The system models eight hospital departments, 156+ simulated patients, 91 beds, and 45 staff across configurable crisis scenarios. Experimental evaluation demonstrates sub-200ms API latency, real-time WebSocket synchronization, and explainable statistical forecasts. MedTwinX demonstrates that responsible healthcare AI requires governance-by-design.
