GeoPython2019

Getting Data out of CAD and into Python
2019-06-26, 09:00–09:30, Room 2

The ad-hoc export of site data and project layout geometry from CAD/BIM teams, together with its organisation and dissemination, requires significant time and effort. Our work has explored and compared manipulating traditional CAD export formats such as DXF, using the BIM IFC data schemes and enabling exchange using concepts from W3C (Semantic Web) Linked Data initiatives.


We have proposed and developed the use of decision support systems (DSS) for civil engineering design in energy and transportation infrastructure projects. Decisions are informed by physical site conditions, including topographic, geologic and hydrologic data, for project layout, data analysis and engineering calculations.

To accomplish this, practical procedures and workflows are needed to exchange data between CAD/BIM (Computer Aided Design / Building Information Modelling) platforms (used for design development and construction drawings) and data analysis programs developed externally to them. Our recent work has involved developing procedures to improve data acquisition from CAD/BIM teams and its subsequent use in Python or more generally on GIS platforms. Experience gained so far shows that DXF export from CAD systems with conversion to GIS raster and vector formats remains a viable (and sometimes the only) approach. We have been using tools available in the Python environment (Pandas, Grass, etc) to accomplish this.

It is however clear from the initial use of our DSS methods that ad-hoc export of site data and project layout geometry from the CAD/BIM teams, together with its organisation and dissemination, requires significant time and effort. This problem will only increase with the amount and variability of data being accumulated from, for example: satellites, drones and field sensors.

The use of BIM IFC data schemes is growing and has the potential to improve the quality of data exchange, with respect to identification, transformation and versioning of data sets. However this can only be achieved with the development of common vocabulary and naming conventions that can be used and shared by design teams.

The work has explored and compared manipulating traditional CAD export formats such as DXF, using the BIM IFC data schemes and enabling exchange using concepts from W3C (Semantic Web) Linked Data initiatives.

Linked Data and associated methodologies and standards (e.g. Resource Description Framework) has the potential to greatly aid in data exchange by providing an existing, documented, open protocol, together with existing tools and methodologies developed in other scenarios for the construction of ontologies (vocabularies) and naming systems and the means by which to apply them to our sphere of interest.