You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After using Planwise in several projects it was pretty clear that the quality of the region shapes (polygons) are not aligned with the quality (level of detail) of the population and accessibility (friction) sources. In other words, we have super detailed region shapes combined with more coarse population and friction rasters.
We might consider using the mapshaper -simplify command to get a balanced quality of the 3 sources.
Here's an example of size for simplification
The main issue with having a highly detailed region shape is that it makes operating with it (eg. for filtering of providers, clipping the demand rasters) very slow while not providing any actual value.
From the slack conversation
The most complicated part would be adding a simplification step when importing the regions and choose some criteria for the precision cut.
We have a size criteria implemented here which is used for the previews. I would prefer a resolution criteria for this use case, and I'm sure it should be easily doable. Just need to stitch it together.
It may be a good time to migrate the remaining Bash scripts to Babashka. Or at least those involved here: load-regions and load-friction-raster.
The text was updated successfully, but these errors were encountered:
We will want to "pair" this with adding a bit of extra border to the friction rasters (using -cblend 1 when cutting the regions in the load-friction-raster script) which should eliminate issues when there are providers near the limits of the regions and they are inside the region shape but they have no friction data.
After using Planwise in several projects it was pretty clear that the quality of the region shapes (polygons) are not aligned with the quality (level of detail) of the population and accessibility (friction) sources. In other words, we have super detailed region shapes combined with more coarse population and friction rasters.
We might consider using the
mapshaper -simplify
command to get a balanced quality of the 3 sources.Here's an example of size for simplification
planwise/scripts/tools/update-region-previews
Line 59 in 7a10252
The main issue with having a highly detailed region shape is that it makes operating with it (eg. for filtering of providers, clipping the demand rasters) very slow while not providing any actual value.
From the slack conversation
The text was updated successfully, but these errors were encountered: