AI enriched Digital Twins for large-scale renovation analysis

AI enriched Digital Twins for large-scale renovation analysis

The DTCC associated project DecarboAIte is applying machine learning (ML) to extract features from publicly available databases to enrich urban digital twin and provide optimized renovation measures for decision-support.

The building sector has a huge potential to save greenhouse gas emissions. Large scale renovations are urgently needed to meet Sweden’s climate goal for 2045. One barrier is the manual, time consuming planning of renovation measures. Here, urban digital twins can provide a good basis for efficient renovation planning on a large scale. The goals are twofold. Firstly, the team is developing a ML-based method to extract information needed to simulate the performance of buildings (3D geometries, windows, material properties). Secondly, an optimization method based on Genetic Algorithms will be developed that includes energy simulation, Life Cycle Assessment and a Life Cycle Cost Analysis.

The project is led by Chalmers Industriteknik and is funded by Vinnova.


Chalmers Industriteknik (Coordinator)

Chalmers University of Technology

GATE institute, Sofia University

ETH Zurich



Energikontor Väst

Helsingborg kommun


EKSTA Kungsbacka

For more information please contact:
Bernd Ketzler,