[Back]

A. Mahdavi, F. Tahmasebi:
"Predicting people's presence in buildings: An empirically based model performance analysis";
Energy and Buildings, 86 (2015), 349 - 355.

tBuilding performance is influenced by occupants´ presence and actions. Knowledge of occupants´ futurepresence and behaviour in buildings is of central importance to the implementation efforts concerningpredictive building systems control strategies. Specifically, prediction of occupants´ presence in officebuildings represents a necessary condition for predicting their interactions with building systems. In thepresent contribution, we focus on the evaluation of a number of occupancy models to explore the poten-tial of monitored past occupancy data towards predicting future presence of occupants. Towards thisend, we obtained long-term high-resolution monitored occupancy data from a number of workplaces(in open, semi-open, and closed office settings) in a university building. Using this data, we trained twoexisting probabilistic occupancy models and an original non-probabilistic occupancy model to predictthe occupancy profiles of the same workplaces on a daily basis. The predictions were evaluated via com-parison with monitored daily occupancy profiles. To conduct the model evaluation in a rigorous manner,separate sets of data were used to train and evaluate the models. A set of five specific evaluation statisticswas deployed for model comparison. In general, the obtained level of predictive accuracy of all modelsconsidered was found to be rather low. However, the proposed non-probabilistic model performed betterin view of short-term occupancy predictions. The results thus facilitate a discussion of the potential andlimitations of predicting building occupants´ future presence patterns based on past monitoring data.

none - see english version.

Occupancy models, probabilistic methods, non-probabilistic methods, predictive building systems control, model evaluation, evaluation statistics

http://dx.doi.org/10.1016/j.enbuild.2014.10.027