The application of building retrofitting solutions targeting improved energy efficiency and thermal comfort is significantly influenced by environmental and climate conditions. This study aims to automate a reliable dataset and enhance the predictability of the post-retrofit performance of the buildings. The proposed hybrid methodology utilises physics-based and data-driven methods to evaluate a range of retrofitting scenarios across diverse UK climate zones and validates an automated dataset with real-time data collected via IoT (Internet of things)-based sensors. This hybrid method enables a comprehensive assessment of retrofitting solutions’ impacts on building performance. The collected data create a reliable dataset and serve as the foundation for training machine learning (ML) prediction models and support decisions in retrofit strategies. The findings reveal that in cool–humid climates, the air source heat pumps significantly perform better when compared to 58 heating systems in terms of the balance of energy efficiency and thermal comfort. Moreover, Water Source Heat Pumps (WSHPs) are recommended for colder regions. As a result, zone-specific retrofitting strategies with seasonal adjustments are recommended for achieving optimum energy efficiency and thermal comfort.

A large body of research has been developed with the aim of assisting policymakers in setting ambitious and achievable environmental targets for the retrofit of current and future building types for energy-efficiency and in creating effective retrofit strategies to meet these targets. The aim of this research is to conduct a comprehensive study to identify the relationship between building typology and sustainability, with a particular emphasis on retrofitting and try to identify research gaps in the most effective energy-saving strategies for retrofitting various types of buildings. In this regard, this study conducts a systematic literature review (SLR) utilizes artificial intelligence (AI) and natural language processing (NLP). Sixty relevant papers are selected and reviewed, establishing a comprehensive searching scheme. The research highlights retrofitting strategies for improving energy efficiency in buildings and discusses the limitations of current practices in terms of physical and technical developments, such as building retrofit assessment according to the typology of the building and environmental factors. To address these limitations, this study proposes a methodology for future research with a focus on in-depth building classification, developing tailored retrofitting alternatives, and establishing an adaptive solution framework. This framework aligns cohesively with diverse typologies, adapts to changing environments, and enhances long-term energy-efficient performance. It proposes detailed building categorization to understand the interconnections between a building's physical characteristics, technology, and energy needs. Additionally, it suggests tailoring retrofit solutions for diverse building types and creating an adaptable framework for changing conditions. Using qualitative research, literature review, quantitative analysis, and case studies, the methodology ensures research credibility. Prototyping is employed to refine processes, considering building types and environmental factors.

This research addresses two specific knowledge gaps. The first regards the influence of domestic low-carbon technology (LCT) installation approaches and occupier status on user acceptance. The second is to demonstrate the role of machine learning techniques in producing an enhanced model-based understanding of domestic LCT acceptance. Together, these two approaches provide new insights into LCT acceptance through the theory of planned behaviour and demonstrate the value of machine learning for modelling such acceptance. Our aim is therefore to contribute to model-based knowledge about the acceptance of domestic LCTs. Specifically, we contribute new knowledge of the acceptance of LCTs according to the theory of planned behaviour and of the value of machine-learning techniques for modelling this acceptance. Through empirical research using an online quasi-experiment with 3813 English residents, we developed a model of low-carbon technology adoption and evaluated machine learning for model analysis. The design factors were the installation approach and occupier status, with main outcomes including adoption intention, willingness to accept, willingness to pay, attitude, subjective norm, and perceived behavioural control. To examine residents’ technology acceptance, we created two virtual reality models of technology implementation, differing in installation approach. For machine learning analysis, we employed nine techniques for model validation and predictor selection: linear regression, LASSO regression, ridge regression, support vector regression, regression tree (decision tree regression), random forest, XGBoost, k-NN, and neural network. LASSO regression emerged as the best technique in terms of predictor selection, with (near-)optimal model fit (R² and MSE). We found that attitude, subjective norm, and perceived behavioural control significantly predicted the intention to adopt low-carbon technologies. The installation approach influenced willingness to accept, with higher intention for new-build installations than retrofits. Homeownership positively predicted perceived behavioural control, while age negatively predicted several outcomes. This study concludes with implications for policy and future research, a specific emphasis upon contemporary UK policy towards Future Homes Standards, and public information campaigns targeted to specific demographic user groups. This research demonstrates the value of an extended theory of planned behaviour model to study the acceptance of LCTs and the value of machine learning analysis in acceptance modelling.