Data Mining of Occupant Behavior in Office Buildings

AbstractLiterature studies confirm occupant behavior is setting the direction for contemporary researches aiming to bridge the gap between predicted and actual energy performance of sustainable buildings. Using the Knowledge Discovery in Database (KDD) methodology, two data mining learning processes are proposed to extrapolate office occupancy and windows’ operation behavioral patterns from a two-years data set of 16 offices in a natural ventilated office building. Clustering procedures, decision tree models and rule induction algorithms are employed to obtain association rules segmenting the building occupants into working user profiles, which can be further implemented as occupant behavior advanced-inputs into building energy simulations