This work studies how environmental conditions, such as carbon dioxide concentration, temperature, and humidity, affect tenants' behaviors in residential buildings, particularly focusing on the window interactions. The data is collected from the KTH Live-In Lab, equipped with state-of-the-art sensors, and whose residents are students. To evaluate heterogeneity in the behavior of each tenant, four apartments and two winters (2019 and 2023) are analyzed separately. In particular, multivariate logistic regression models are built to estimate the probability that a window is opened or closed at a certain time. Our results show that the time of the day and indoor temperatures are often linked to window-related actions. These efforts work towards integration into control systems of smart buildings to improve buildings’ energy efficiency and lead to more sustainable housing for a broader population.
Bukorovic et al. (Wed,) studied this question.