Stochastic behavioral models of bedroom window operation in sub-tropical residential buildings

Abstract

This study investigates window opening and closing behaviors in residential bedrooms in Hong Kong by examining occupants' habits regarding window operation, window opening and closing behavior, and the influence of environmental and contextual factors. Using self-reported survey data on window opening and closing behavior and outdoor environmental data from an open-source database, univariable and multivariable logistic mixed-effect models were developed to evaluate the impact of various parameters. The findings indicate that outdoor environmental factors have a limited influence on window opening and closing probabilities. Conversely, time of the day was identified as a significant factor, emphasizing the role of occupant habits and preferences in determining these probabilities. Importantly, the study results agreed with the observations that the drivers between window opening and window closing differed, further emphasizing the need for separate modeling. The study highlights the importance of considering environmental and contextual factors when predicting window operation behaviors. This study contributes to our understanding of building occupant behavior and provides valuable insights for building designers and engineers to enhance energy simulations and optimize building performance.