A data-driven retrofit framework for historic buildings: the case study of Villa Farnesina in Rome

Abstract The energy retrofit of buildings represents a fundamental challenge in meeting national and European climate objectives, particularly in Italy, where a significant portion of the building stock was constructed prior to the introduction of energy efficiency regulations. This study focuses on energy retrofit of historical buildings, which are typically characterized by complex regulatory and technical barriers, making it crucial to apply interdisciplinary and data-driven solutions. To highlight these aspects, this work presents the study of the energy performance of Villa Farnesina, a 16th-century heritage building located in Rome, by using dynamic energy simulation calibrated against real consumption data. The model validation yielded a Normalized Mean Bias Error (NMBE) of 0.4 % and a Coefficient of Variation of the Root Mean Square Error (CV(RMSE)) of 16.5 %, confirming the model’s reliability, while highlighting the need for improved input data on occupancy and system operation.