With the “Mesh Inclusion” component you are making a datatree matching error.
You are using “Point in Breps” component for breps (notice the plural “s”), but “Mesh inclusion” for the meshes.
“Mesh inclusion” works with a single mesh at time.
In each of your point tree branches the component is testing the first point with the first mesh, second with second, etc etc, and then using the last mesh for all the points until the end (280).
Your points are in an already complex datatree, and making the proper matching to test inclusion to all the meshes is needlessly even more complex.
You can approach this in two ways, and use both at the same time:
- flatten your point tree (and later unflatten it)
and/or - join the meshes into a single mesh
Problem: as your breps have some congruent edges, the resulting mesh will have some non-manifold edges, making it fail completely any point inclusion test.
So, first do a solid union of your breps, and only then convert it into a mesh.
About the optimization of the testing, maybe this is related/useful: Subdivided Voxelization:
If possible, explain how your point tree is structured.
Optimization might require starting with larger cells and then sub-sampling.