Currently some identified superblocks are either unrealistically large or small. Eggiman is solving this by filtering for some criteria like street lengths, see: https://www.nature.com/articles/s41893-022-00855-2/tables/1
A future feature could implement a similar postprocessing step after the partitioning step of pruning or filtering, or a merging/breaking up of too small/large superblocks. Related, a LTN plausibility scoring system: https://www.sciencedirect.com/science/article/pii/S2950105925000415
Currently some identified superblocks are either unrealistically large or small. Eggiman is solving this by filtering for some criteria like street lengths, see: https://www.nature.com/articles/s41893-022-00855-2/tables/1
A future feature could implement a similar postprocessing step after the partitioning step of pruning or filtering, or a merging/breaking up of too small/large superblocks. Related, a LTN plausibility scoring system: https://www.sciencedirect.com/science/article/pii/S2950105925000415