Smart restroom: preventive maintenance system using machine learning algorithm/
Charlin M. Infante, Ryan Rey D, Magdalita, Joey Boy E. Mission, John Michael P. Pontanos, and Rafaela E. Santos.--
- Manila: Technological University of the Philippines, 2025.
- xii, 168pages: 29cm.
Bachelor's thesis
College of Industrial Technology.--
Includes bibliographic references and index.
Traditional restroom maintenance practices are predominantly reactive, leading to inefficiencies, increased operational costs, and user dissatisfaction due to unexpected equipment failures. This thesis aimed to develop the "Smart Restroom," a preventive maintenance system leveraging IoT sensors and machine learning (ML) algorithms to transition from reactive to proactive facility management. The system captures critical parameters predictive of maintenance needs by integrating sensors to monitor real-time data such as usage frequency, environmental conditions, and consumable supply levels. Machine learning models, employing anomaly detection and time series forecasting, analyze this data to anticipate equipment malfunctions and optimize maintenance schedules. The proposed framework not only predicts imminent failures but also prioritizes tasks based on urgency, enhancing resource allocation. Experimental results demonstrate the system’s efficacy in reducing downtime, lowering maintenance costs, and improving hygiene standards through timely interventions. Furthermore, this study highlighted the system’s scalability and adaptability, positioning it as a viable component of smart city
infrastructures aimed at sustainable urban management. The innovation lies in its data- driven approach, which transcends conventional threshold-based alerts, offering dynamic
insights tailored to specific restroom usage patterns. This thesis underscored the transformative potential of IoT and ML in public infrastructure, advocating for their broader adoption to enhance operational efficiency and user experience in communal spaces.