Digital twin-enabled built environment sensing and monitoring through semantic enrichment of BIM with SensorML

Abstract

Effective environmental condition monitoring provides constant surveillance of the built environment and reveals deteriorations that could impact the daily operation of facilities, especially amid COVID situations. However, the current Industry Foundation Classes (IFC) data schema for Building Information Modelling (BIM) provided limited support to represent full semantics related to environmental sensing and monitoring. How to semantically enrich the IFC schema with enhanced data description capability for informed decision-making in smart facilities management (FM) amid COVID situations remains an open question. This paper develops a semi-automatic extension and integration of IFC data schema with Sensor Model Language (SensorML) specification in order to support automated built environment sensing and monitoring. Referring to SensorML, an extended IFC model view definition for a comprehensive description of required sensor metadata and sensing entities is presented. An Internet of Things (IoT) sensor network is then established to realise continuous data collection from a variety of wireless sensing devices. The spatial-temporal data captured by the IoT sensor network are extracted by a regular expression-based data distillation algorithm and integrated with the digital twin, in which spatial interpolation algorithms further analyse, compute, and visualise the state of the environment. The proposed method is demonstrated via an experimental study which supports real-time environmental monitoring and delivers more actionable insights to facility managers to sustain the daily operation of buildings. This study contributes new methods and models to semantically enrich the digital twin from the data perspective for environmental condition monitoring during the pandemic time which fosters the development of holistic building facility management.