The RESILIENT project has developed an automated method for modelling national-scale reference buildings

The RESILIENT project has developed a novel automated approach for identifying and modelling national-scale reference buildings, demonstrated through a case study on the potential of rooftop-based interventions in Spain. The results have been published in the Journal of Building Engineering (JCR-Q1, Impact Factor: 7.2) in September 2025, titled “An automated approach for modelling national-scale reference buildings: A case study on the potential of rooftop-based interventions in Spain.” The study was conducted by PhD. candidate Juan José Medina-Musellas, under the supervision of Prof. Marta Gangolells and Prof. Miquel Casals, with contributions from the PhD. Nicolás Carrasco-Astudillo and Dr. Kàtia Gaspar, all members of the Group of Construction Research and Innovation (GRIC) at the Universitat Politècnica de Catalunya (UPC).

Accurate modelling of building energy performance is essential for transforming existing buildings into climate-resilient structures. However, the modelling of entire building stocks requires extensive data and computational resources. This paper presents a fully automated method that integrates data merging, clustering, geometric modelling, and energy performance characterization to efficiently identify and model representative buildings. The method combines energy performance certificates and 3D cadastral data to generate detailed energy models that accurately reflect national building typologies.

The approach was applied to the Spanish residential stock to evaluate the potential of rooftop-based retrofit strategies such as: green roofs, photovoltaic systems, and rainwater harvesting, to enhance climate resilience. Using the k-means clustering algorithm, the study identified 14 reference buildings that represent the most common residential typologies in Spain, differentiating between sloped and flat roof configurations.

This new framework provides a scalable, replicable, and data-driven tool that significantly reduces modelling time and computational effort, while supporting policy development and decision-making for sustainable building renovation at the national level.

The methodological workflow of the automated modelling process.

You can read the full text of the article at: https://doi.org/10.1016/j.jobe.2025.114014