Through the use of a data-centric technique for analyzing Building Man- agement System (BMS) data, the article discusses the shortcomings of current build- ing data acquisition for capturing occupants’ behaviour. Applying machine learning to a real 3-years dataset, the potential of unsupervised learning for discriminating between normal daily patterns (motifs) and abnormal ones (discords) for automatic fault detection is outlined, as well as the difficulty to extract from typical BMS data meaningful insights about usage. Finally, the authors propose design guidelines to better monitor, e.g. acquire and analyze, building function together with building usage through digital means, identifying BMS data as a candidate vehicle for this purpose.
CITATION STYLE
Nembrini, J., Sánchez, R., & Lalanne, D. (2020). Discussing the Potential of BMS Data Mining to Extract Abnormal Building Behaviour Related to Occupants’ Usage. In Impact: Design With All Senses (pp. 727–736). Springer International Publishing. https://doi.org/10.1007/978-3-030-29829-6_56
Mendeley helps you to discover research relevant for your work.