Skip to main content

Hello, I have the following problem and hope you can give me a tip.

For a statistical analysis of my city, I have to know how much building space is on a landparcel. For this I have a source with the parcels and one for the buildings.
I wanted to solve the problem with the AreaOnAreaOverlayer Transformer. Now I have the problem that some buildings are on more than one parcel. In this case, the common parcels must be merged and all buildings identified.

Here's an example:

The Buildung 1 is located on the parcels A,B,C. Buildung 2 is located on the parcels B,D.

At the end the parcels A,B,C,D must be dissolved.

When i use the AreaOnAreaOverlayer with Buildings and Landparcels, i have a lot of areaparts with several combinations of building and parcel-ID. When i start to dissolve the areas again with the id of one common building, i loose IDs of the others.

Please can you give me a tip how this problem can be solved? Thanks, David

How about using the Clipper:

  1. Calculate "before" parcel area (AreaCalculator)
  2. Clip against buildings (Clipper)
  3. Calculate "after" area (AreaCalculator)
  4. Calculate difference or percentage between "before" and "after" parcel area (ExpressionEvaluator)


Thank you for your prompt reply.

My problem is that if a building overlaps several parcels, i have to calculate the metric for all buildings on all affected parcels. So I have to dissolve the green marked area in my example.


Thank you for your prompt reply.

My problem is that if a building overlaps several parcels, i have to calculate the metric for all buildings on all affected parcels. So I have to dissolve the green marked area in my example.

I'm not quite sure I understand exactly what you need. Did you try the Clipper method and what was it about it that didn't meed your requirements?

If I understood the requirement correctly, the ListCombiner (from FME Hub) in this context might help you.



Thank you, takashi and david. The solution of takashi is exactly what i need. That helped me alot.


Reply