Everyone, I'm trying to use the balanced zones tool to create hydrant inspection areas. These used to be drawn by hand but I'm thinking this tool could help us automate a lot of this process.
I have the following features:
The criteria are that we want the zones to have roughly equal number of hydrants within them, and want the distribution around the stations to be equal as well. One complicating factor with the stations though is that they have different staff levels. This is represented by an integer column, 1-3. So for a station with a 3, we'd want that to be responsible for inspecting 3x the number of hydrants as a station with a 1.
One way I could do this, at least in part, is to use defined number of zones in the creation method, using the hydrant points. But the output isn't ideal because its points that I then have to convert into polygons. I can do this with aggregate points or minimum bounding area, but that doesn't produce polygons with coincident edges.
This video shows how to build zones using existing polygons. I could try to emulate this with my existing HMZ polygon. It has a column for number of hydrants in the zone, and I could also add 'station in zone' 0/1 as a column. But the zones are much bigger than the polygons in the example video, and I'm still not seeing how to take the staffing levels in account. And another potential monkey wrench is the street network, we don't really want a zone to be split in the middle by a highway with no nearby onramp, that sort of thing.
I'm fine with having to run multiple tools, or do some steps of this manually. Right now I'm not seeing how to do this given the constraints. This must be at least a somewhat common task for local government given the data. Since I know the study area boundaries, the number of hydrants, and stations, even if I could just produce polygons covering each station area, taking the staffing levels into account, that would get me halfway there and I could split them up manually after that.
Any ideas or thoughts on which tool I should use? Thanks.
Did you ever find a good way to do this? Looking to do something similar myself.
Hi Matthew, unfortunately no. Just doesn't seem possible to make it work with our data. Complex geography + variables on the data means I just couldn't get good results.
We ended up redrawing the boundaries using the existing inspection areas.