Creating realistic geometry Out of OSM data

edit

@Avi, that is a good question. Yes there is a way to do this, at least for constructions that don't use No Mass material, but I'm not aware of any measures currently setup to do this. You can get a lot of the same benefits by instead using frame and divider for windows to account for the wall thickness at the windows. Also for drop ceilings you could add an interior partition surface or model a plenum zone; this could be useful if you are using using a daylighting strategy that the ceiling position is important for.

One issue you have to address if you want to export a 'thick wall' model from an OSM or IDF is to know the assumptions were in the energy model related to the placement of the thin surface planes. A general practice is for exterior surfaces to be drawn to the outside edge of the construction, while interior surfaces are drawn to the centerline. This is more true for walls than floor and roofs. Generally I see floors and roof surfaces drawn at the top edge of the construction (ground slab and roof included). This affects the interpretation of the height of the window above the floor and placement of daylight sensors. If you do make a measure like this it would be good to define the expectations about thin surface placement so people can produce models that will be properly interpreted by your measure. You could also expose some arguments that give more control.

Depending on what tool you are producing the model for it might be easier to use their API. For example walls in Revit are drawn on a reference plane within a construction that could be at the centerline, outside edge, face of stud, etc. You can let Revit handle the thick wall extrapolation of thin wall geometry.

Just for reference here are a few relevant posts. https://unmethours.com/question/20758... https://unmethours.com/question/8724/...