Efficient inventory and equipment maintenance management in university medical laboratories poses a significant technical challenge, primarily due to the frequent reliance on manual record-keeping methods, which lead to data inaccuracies and hinder the planning of preventive and corrective maintenance. This article proposes a modular architecture specifically designed for the automated management of inventory and predictive maintenance in university medical laboratories. The architecture is composed of four key modules: (1) data acquisition, (2) indicator analysis, (3) dynamic threshold management, and (4) interactive visualization. It integrates a dynamic monitoring model that employs adaptive thresholds to adjust maintenance strategies in real time, based on key performance indicators such as Mean Time Between Failures (MTBF) and Mean Time to Repair (MTTR). The system was validated in a real-world setting at Universidad Politécnica Salesiana, managing 479 medical devices across various laboratories. The quantitative results show a 25% reduction in downtime, a 15% increase in MTBF, and a 20% decrease in MTTR. Additionally, the system achieved 90% prediction accuracy, calculated as the percentage of alerts that correctly matched confirmed failure events during the validation period, significantly outperforming traditional maintenance methods.