A predictive model for an effective maintenance of hospital critical systems

Abstract

This paper presents a predictive model for the maintenance of critical systems in hospital facilities. The developed model is based on machine learning algorithms and data acquired from the Building Management System (BMS) and supported by the Computerized Maintenance Management System (CMMS). Support Vector Machine (SVM) and Prophet forecasting algorithms are used to assess the current condition of the system and to predict its future conditions. The model was applied to Air Handling Units (AHU) of the Heat Ventilation and Air Conditioning (HVAC) system of a hospital in Jordan. The AHU is considered one of the critical systems in the hospital as it is responsible for controlling the Indoor Air Quality(IAQ) of the building. The developed model achieved an acceptable accuracy in both current condition assessment and future condition prediction. The study has also highlighted the benefits of implementing the model to the hospital in terms of increasing the effectiveness of HVAC system operation and maintenance and cost reductions. The model is set to be integrated with advanced monitoring and maintenance technologies to optimize the performance of the hospital critical systems.