Hi all,
I’m sharing a publication and code for a GIS-to-UBEM workflow based on 24 different building archetypes, using Grasshopper with the Urbano and Dragonfly plug-ins.
The UBEM also incorporates WWR values for each orientation of each building (for a total of 1,432 building volumes), detected using a deep learning model.
Article title: “Detecting window-to-wall ratio for urban-scale building simulations using deep learning with street view imagery and an automatic classification algorithm”.
Link to the article: https://doi.org/10.1007/s12273-025-1301-3
Link to the GitHub repository: https://github.com/asuptor/WWRdetection
Cheers,
Anthony