Reshaping Smart Energy Transition: An analysis of human-building interactions in Qatar Using Machine Learning Techniques

TL;DR

This study uses machine learning to analyze human behaviors and perceptions influencing energy use in Qatar's building sector, providing insights for tailored energy policies and promoting sustainable development.

Contribution

It introduces a machine learning-based analysis of socioeconomic and behavioral factors affecting energy consumption and comfort in Qatar, focusing on diverse expatriate communities.

Findings

Human factors significantly influence comfort perception and energy habits.

Socioeconomic and behavioral dimensions correlate with energy use discrepancies.

Implications for targeted energy policy interventions and social well-being improvements.

Abstract

Policy Planning have the potential to contribute to the strategic development and economic diversification of developing countries even without considerable structural changes. In this study, we analyzed a set of human-oriented dimensions aimed at improving energy policies related to the building sector in Qatar. Considering the high percentage of expatriate and migrant communities with different financial and cultural backgrounds and behavioral patterns compared with local communities in the GCC Union, it is required to investigate human dimensions to propose adequate energy policies. This study explored the correlations of socioeconomic, behavioral, and demographic dimensions to determine the main factors behind discrepancies in energy use, responsibilities, motivations, habits, and overall well-being. The sample included 2,200 people in Qatar, and it was clustered into two consumer categories: high and low. In particular, the study focused on exploring human indoor comfort perception dependencies with building features. Financial drivers, such as demand programs and energy subsidies, were explored in relation to behavioral patterns. Subsequently, the data analysis resulted in implications for energy policies regarding interventions, social well-being, and awareness. Machine learning methods were used to perform a feature importance analysis to determine the main factors of human behavior. The findings of this study demonstrated how human factors impact comfort perception in residential and work environments, norms, habits, self-responsibility, consequence awareness, and consumption. The study has important implications for developing targeted strategies aimed at improving the efficacy of energy policies and sustainability performance indicators.