This study proposes and evaluates an IoT-enabled system for managing medical equipment assets throughout their lifecycle in a smart hospital setting. The framework adds real-time tracking, predictive analytics, and optimization to biomedical engineering processes and addresses underutilization, unexpected downtime, and scattered maintenance practices. A major hospital deployed multi-sensor nodes on high- and medium-criticality devices like ICU monitors, ventilators, and mobile imaging units, streaming usage, health, and environmental data to an analytics platform. The framework was implemented and evaluated over a 12-month period in a tertiary-level smart hospital that manages approximately 3,500 medical equipment assets across intensive care units, operating theatres, radiology departments, and general wards. We employed a stochastic degradation process and Markovian transitions to model lifecycle states (healthy, degraded, at-risk, failed, retired). To determine the remaining useful life and failure risk, we used a hybrid prediction ensemble that combines tree-based models and LSTM networks. These results were sent to a cost-based optimization tool that generated risk-aware repair and replacement plans. In the IoT + optimization setup, yearly downtime and major failures were reduced significantly compared to standard practice and IoT-monitoring-only scenarios. Average fleet uptime increased, and lifecycle maintenance cost decreased. Spatial and association analyses confirmed links between use, environment, and degradation, and sensitivity studies showed the optimization remained strong despite parameter error. The results show IoT-driven lifetime intelligence can shift equipment management from reactive, schedule-driven care to proactive, data-driven asset ownership.
Yu et al. (2026) studied this question.