A Digital Twin and IoT-Enabled Predictive Maintenance and Energy Optimisation Framework for Hospital Wastewater Treatment Systems

• S. Embarki Applied Sciences Research Laboratory, Mohammed First University, Oujda, Morocco https://orcid.org/0000-0002-4365-1562
• Y. El Kihel Laboratoire d’Innovation Numérique pour les Entreprises et les Apprentissages au service de la Compétitivité des Territoires, LINEACT-CESI, Bordeaux, France https://orcid.org/0009-0000-9600-3908
• B. El Kihel Applied Sciences Research Laboratory, Mohammed First University, Oujda, Morocco https://orcid.org/0000-0001-8628-0397

Abstract

This study develops an integrated digital twin framework for electrical optimisation and predictive maintenance in hospital wastewater treatment systems. From an industrial engineering perspective, the framework links real-time monitoring, load scheduling, power-quality control, condition assessment, and maintenance planning in a single decision-support architecture. Physical measurements from IoT sensors are synchronised with virtual electrical and process models so that facility managers can evaluate operating states, identify energy inefficiencies, and prioritise interventions. Optimisation combines power factor correction, harmonic mitigation, peak demand reduction, and variable-frequency-drive control, supported by machine learning. Predictive maintenance combines electrical, thermal, and vibration indicators to assess equipment condition. Validation on a laboratory-scale pilot system showed sustained power-quality improvements, a 21.7% reduction in energy consumption, and a 22% reduction in peak demand. These gains corresponded to annual savings of 22,107 MAD and a simple payback period of 3.8 years. The results demonstrate the practical value of the proposed framework for energy-intensive systems, while further full-scale validation is required before generalisation to hospital facilities with different layouts and duty cycles.