Monitoring in Operation and Maintenance

8 Assets play a significant role in delivering the functionality and serviceability of the building 9 sector. However, there is a lack of efficient strategies and comprehensive approaches for 10 managing assets and their associated data that can help to monitor, detect, record, and 11 communicate operation and maintenance (O&M) issues. With the importance of Digital Twin 12 (DT) concepts being proved in the architecture, engineering, construction and facility 13 management (AEC/FM) sectors, a DT-enabled anomaly detection system for asset monitoring 14 and its data integration method based on extended industry foundation classes (IFC) in daily 15 O&M management are provided in this study. Following the designed IFC-based data structure, 16 a set of monitoring data that carries diagnostic information on the operational condition of 17 assets can be extracted from building DTs firstly. Considering that assets run under changing 18 loads determined by human demands, a Bayesian change point detection methodology that 19 handles the contextual features of operational data is adopted to identify and filter contextural 20 anomalies through cross-referencing with external operation information. Using the centrifugal 21 pumps in the heating, ventilation and air-cooling (HVAC) system as a case study, the results 22 indicate and prove that the developed novel DT-based anomaly detection process flow realizes 23 a continuous anomaly detection of pumps, which contributes to efficient and automated asset 24 monitoring in O&M. Finally, future challenges and opportunities using dynamic DTs for O&M 25 purposes are discussed. 26