Use of advanced digital methodologies for the analysis of complex systems. HBIM for the development of cognitive/energy‑responsive digital twins

This study proposes an integrated framework for developing Cognitive Digital Twins (CDT) to transform heritage buildings into proactive nodes within smart grid infrastructures. By the integrated implementation of HBIM, IoT sensors, Machine Learning and Lean methodologies could be possible to balance energy efficiency with heritage building’s preservation. To achieve this, the framework follows a three-phase pathway: foundational data acquisition through standardized HBIM protocols; semantic algorithm synthesis using Neural Networks and Federated Learning for predictive forecasting; and validation through pilot deployments. The scope of the paper is the analysis of the system’s cognitive capacity like perception, reasoning, and learning, for enabling autonomous adaptation to environmental and occupancy variations of the reference parameters. This could enhance a deep Lean behaviour, where the CDT is the approach that established a clear communication among process actors, reducing wastes related to the lack of interoperability of data and information, and definitively is able to provide information to proactive maintenance phases. This approach aligns heritage conservation with the 2030 Agenda, positioning protected buildings as technologically advanced components of resilient, sustainable smart cities.