CityJSON and Digital City Twins: From OSM to Smart City

2026-02-043 min read

CityJSONSmart City3DBIM

Export formats: CITYJSON CITYGML IFC GLTF

A digital twin of a city is a 3D model connected to real-world data: building energy consumption, traffic flows, air quality, noise levels. Standards like CityGML and CityJSON form the foundation of these models. OpenStreetMap is one of the best data sources for creating them.

CityJSON: Compact Standard for 3D Cities

CityJSON is a JSON implementation of the CityGML standard, developed by the Open Geospatial Consortium (OGC). Unlike CityGML (XML), CityJSON is 6-10x more compact and more developer-friendly.

Key elements of CityJSON:

CityObjects — buildings, roads, bridges, trees, water bodies, terrain.
Geometry — 3D geometry of objects: Solid, MultiSurface, CompositeSurface.
Semantics — semantic surfaces: roof (RoofSurface), wall (WallSurface), floor (GroundSurface).
Attributes — arbitrary attributes: construction year, material, function.
LOD (Level of Detail) — detail levels from LOD0 to LOD4.

{
  "type": "CityJSON",
  "version": "2.0",
  "CityObjects": {
    "building_001": {
      "type": "Building",
      "attributes": {
        "yearOfConstruction": 1952,
        "function": "residential",
        "measuredHeight": 24.5
      },
      "geometry": [{
        "type": "Solid",
        "lod": "2",
        "boundaries": [[[0, 1, 2, 3], [4, 5, 6, 7]]]
      }]
    }
  },
  "vertices": [[37.618, 55.751, 0], [37.619, 55.751, 0], ...]
}

Levels of Detail (LOD)

CityJSON supports 5 levels of detail:

LOD	Description	Typical Application
LOD0	2.5D — building footprints with height	Overview maps, statistics
LOD1	Extrusion — "boxes" with flat roofs	Visualization, shadowing
LOD2	Roof shapes, walls, bases	Solar energy, noise maps
LOD3	Doors, windows, balconies, textures	Architectural design
LOD4	Interiors, furniture, utilities	BIM, facility management

OpenStreetMap contains data for LOD0-LOD1 (building footprints + heights from height and building:levels tags). LOD2 and higher require LiDAR data or BIM models.

Digital Twin Applications

Solar Energy

A LOD2 city model allows calculating solar insolation for every roof. This is critical for solar panel planning. Software: Ladybug Tools, Solar3Dcity, Solis.

Noise Modeling

Traffic noise propagates considering buildings as screens. 3D model + traffic flows = noise map according to the CNOSSOS-EU standard. Software: NoiseModelling, CadnaA.

Energy Modeling

A digital twin of a building in CityJSON/IFC contains geometry, materials, glazing. This enables modeling heat loss and energy consumption. Software: EnergyPlus, SimStadt, TEASER.

Urban Planning

Visualization of new construction in the context of existing buildings. Assessment of impact on neighboring buildings' solar access, views, and wind environment.

Exporting 3D Models from OSM

On osm2cdr.ru you can export a city 3D model in several formats:

CityJSON — OGC standard for 3D cities, LOD1 geometry from OSM.
CityGML — XML version of the same standard, for enterprise systems.
IFC — BIM format for integration with Revit, ArchiCAD, Tekla.
glTF — 3D format for visualization in browsers and games.

Models are built from OpenStreetMap data: building footprints (building=*), heights (height, building:levels), building type (building:material, roof:shape).

Create 3D city models for urban planning projects, energy analysis, and Smart City at osm2cdr.ru.