Occupant Behavior of Window Opening and Closing in Office Buildings: Data Mining Approaches

Occupant behavior is stochastic, complex, and multi-disciplinary. Studies have shown significant impact of occupant behavior on energy use and environmental performance of both residential and commercial buildings. The understanding of the relationship between occupant behavior and building energy consumption can be seen as one of the most effective ways to bridge the gap between predicted and actual energy consumption in buildings. However effective methodologies to remove the effects of other variables on building energy consumption and isolate the leverage of the human factor precisely are still poor investigated. Moreover, the use of data mining approaches in finding meaningful correlations in a large data set is rarely discussed in existing literature. In a view of these facts, this study employs two data mining methods, cluster analysis and association rules, to discover patterns of windows opening and closing in a dataset with: 10-minute interval data over two complete years, 16 offices of a natural ventilated building, and a dozen measured indoor and outdoor physical parameters. The windows opening/closing patterns consider diversity and presence of occupants, time of day and day of year, and important indoor and outdoor environmental parameters. The proposed data mining approaches can be used to disaggregate occupant behavior into clusters and to categorize typical drivers of behavior in office buildings. Final goal is to identify valid, novel, potential useful and understandable patterns of occupant behavior into measured building data. The identified windows opening/closing patterns will be represented as typical occupant profiles that can be used as input to current building energy modelling programs, like EnergyPlus and IDA-ICE, to investigate impact of windows opening and closing behavior on energy use and design of natural ventilation in buildings