Answers are grounded in this paper and cite their sources.
This paper provides a state-of-the-art review of the integration of digital twin and deep learning technologies in the context of indoor environmental conditions (IEC). A total of 136 papers published between 2018 and 2024 were systematically selected and analyzed. The review is organized along two dimensions: (1) the IEC perspective, encompassing key environmental factors such as thermal, air quality, acoustic, and lighting conditions; and (2) the technological perspective, focusing on deep learning model types, generative AI approaches, information flow, and training strategies within digital twins. The review is then supported by illustrative examples from the literature that demonstrate this integration through data curation, model selection, deployment, continuous learning, and considerations of safety and privacy. Lastly, the review identifies current gaps in the body of knowledge and outlines future research directions. This paper contributes to the field by addressing deficiencies in existing reviews and establishing a unified understanding of how deep learning enhances digital twin capabilities for predictive, adaptive, and occupant-centric IEC analytics and management. • Reviews 136 studies on deep learning and digital twins for indoor conditions (IEC) • Analyzes digital twin applications in IAQ, thermal, acoustic, and lighting conditions • Analyzes deep learning models, information flow, and training for IEC • Proposes a unified framework linking deep learning with digital twin systems
No takes yet. Share an insight, caveat, or question.
Haghighat et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: