Deep learning analysis on energy efficiency impacts on sustainable built environments: a Re-eVAM approach
Aurora Greta Ruggeri, Rubina Canesi, Giuliano Marella, Laura Gabrielli, Massimiliano ScarpaPurpose
This study investigates the impact of energy efficiency (EE) improvements on residential property values in Italy, aligning with the European Union's energy performance buildings directive IV (EPBD IV). By leveraging advanced AI and deep learning techniques, the research aims to quantify the relationship between energy performance certificates and market values, contributing to the development of smart and sustainable built environments. The study provides crucial insights for professionals, investors, and policymakers, supporting the transition towards more sustainable urban development and climate-responsive architecture.
Design/methodology/approach
The research employs a novel four-phase methodology, integrating big data analytics with deep learning techniques to address the complex challenges of sustainable built environments. It begins with comprehensive data mining of Italian Real Estate listings, followed by statistical analysis. A Residential building energy-efficient Valuation Model (Re-eVAM) is developed using deep artificial neural networks to forecast property values based on energy performances. Finally, scenario modelling evaluates potential market impacts under various EPBD IV implementation strategies, contributing to data-driven optimization for sustainable regulations and governance.
Findings
The study reveals significant potential impacts of EE improvements on RE values in the Italian market. In the worst-case scenario, insufficient implementation of energy retrofits could potentially devalue up to 74% of the overall RE market. Conversely, comprehensive EE improvements could enhance market value by up to 27% in the best-case scenario, demonstrating the economic viability of sustainable building practices. These findings underscore the critical role of well-designed policies in creating smart and healthy living environments, and the potential for significant value creation through EE improvements in the built environment.
Originality/value
This research provides unprecedented insights into the Italian residential RE market's response to EE measures, utilizing a large, previously unexplored database. The innovative Re-eVAM approach offers a new tool for valuing EE in properties, showcasing the potential of machine learning in creating more sustainable and resilient urban environments. By quantifying the potential economic impacts of various EPBD IV implementation scenarios, this study contributes to the development of smart cities and guides sustainable urban investment strategies. The findings demonstrate the importance of data-driven approaches in achieving the UN Sustainable Development Goals in the built environment sector.