Machine-learning to analyze human-building interactions - How do people use mobile solar protections - Article

Aug 5, 2026

Upload by Oliver

Buildings significantly impact urban energy consumption, and mobile passive solutions such as manual solar protections can reduce heat gains, yet their effectiveness depends on occupants’ decisions. Interpreting adjustments of manual systems is difficult because human behavior is complex and varies across time. This study analyzes 359 manually controlled solar protections over one year using systematic photographic data and unsupervised clustering to classify positions, then manual tagging. Results show protections are mostly closed year-round, with limited variation between occupied/non-occupied days and warm/cool seasons, while one-third were never operated.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Introduction

Context: buildings, energy use, and occupant behavior

Simulation approaches for human behavior in buildings

Prior studies and the need for real case understanding

Journal of Building Engineering 98 (2024) 111039 Contents lists available at ScienceDirect Journal of Building Engineering journal [homepage:](homepage: www.elsevier.com/locate/jobe)[ www.elsevier.com/locate/jobe](homepage: www.elsevier.com/locate/jobe) | Machine-learning to analyze human-building interactions: How do people use mobile solar protections?
Marc Roca-Musacha, * , Gloria Serra-Coch b, Isabel Crespo Cabilloa , Helena Coch Rouraa
a Universitat Polit`ecnica de Catalunya, Architecture, Energy and Environment, Research Group (AiEM), Av. Diagonal 649, 08028, Barcelona, Spain b EPFL Laboratory for Human Environment Relations in Urban Systems (HERUS), School of ENAC, ´Ecole Polytechnique F´ed´erale de Lausanne (EPFL), CH-1015, Lausanne, Switzerland | | | | --- | --- | --- | | A R T I C L E I N F O | | A B S T R A C T | | Keywords:
Machine learning Building observation Adaptive façades Solar protection Manual control systems
User interactions with building systems | | Buildings significantly impact urban energy consumption. Mobile passive solutions, such as manual solar protections, can mitigate heat gains, but their effectiveness depends on occupants’decisions. Analyzing occupant adjustments of manual systems is challenging due to the complexity of human behavior. While multiple studies have observed human-building interactions, analyzing large buildings over extended periods remains a technical challenge. This study examines 359 manually controlled solar protections over a year using systematic photographic data. To manage the large volume of images, an unsupervised machine learning algorithm was employed to cluster similar solar protection positions for each window, which were then manually tagged to identify their positions. Results show that solar protections are mostly closed year-round (28 % aperture on average), with minimal differences between occupied and non-occupied days and between warm and cool seasons. Notably, one-third of the solar protections were not operated throughout the year. Our results provide insights into the usage of manual façade systems and offer valuable knowledge for energy simulation. Methodologically, the use of machine-learning algorithms presents a new way to process large datasets of images, emphasizing the importance of prioritizing quality over quantity and strategic data collection. Future research can apply this methodology to different building types and climates to gain broader insights into occupant behavior and solar protection interactions. | 1. Introduction In urban systems, the impact of buildings on energy consumption patterns emerges as a significant factor, influencing not only energy grids but also contributing to phenomena such as the urban heat island. The energy performance of buildings is significantly influenced by the behavior and actions of the occupants. The interaction of users with building’s active and passive systems can lead to important variations in energy consumption and indoor environmental conditions [1–4]. These variations underline the importance of considering human behavior in energy-performance modeling and simulations. However, simulating human behavior within buildings presents substantial challenges due to its inherently complex and unpredictable nature. Unlike mechanical systems, human actions are influenced by a myriad of factors, including personal preferences, cultural habits, and varying degrees of awareness about energy use [5–7]. This complexity poses difficulties in creating accurate and reliable simulations that reflect real-world scenarios. * Corresponding author. E-mail [address:](address: marc.roca.musach@upc.edu)[ marc.roca.musach@upc.edu](address: marc.roca.musach@upc.edu) (M. Roca-Musach). [https://doi.org/10.1016/j.jobe.2024.111039](https://doi.org/10.1016/j.jobe.2024.111039) Received 4 June 2024; Received in revised form 19 September 2024; Accepted 10 October 2024 Available online 11 October 202

FAQs

Why is analyzing occupants’ use of manual solar protections challenging?

Effectiveness depends on occupants’ decisions, and modeling these adjustments is hard because human behavior is complex, variable, and influenced by many factors.

How did the study process the large image dataset?

An unsupervised machine learning algorithm clustered similar solar protection positions for each window, and then researchers manually tagged the clusters to identify positions.

What were the key findings about solar protection usage over the year?

Solar protections were mostly closed year-round on average, with minimal differences between occupied and non-occupied days and between warm and cool seasons; about one-third were not operated during the year.

Aug 5, 2026

Upload by Oliver