Less active travel in Go Active scenario than in baseline

[joeytalbot]

In Tresham (a remote rural site) the Go Active scenario is giving negative numbers of people walking to work.

[Robinlovelace]

Well spotted. Can you share a link to the data file showing it? I'm just looking at https://raw.githubusercontent.com/cyipt/actdev/main/data-small/tresham/desire-lines-few.geojson and cannot see any negative numbers associated with Go Active.

[joeytalbot]

In https://raw.githubusercontent.com/cyipt/actdev/main/data-small/all-sites.geojson percent_drive_convertable and percent_mapped_drive_convertable are both negative numbers, and percent_commute_active_scenario is lower than percent_commute_active_base.

[Robinlovelace]

Interesting. Do you know the cause of that? My understanding was that the percent_* columns in all-sites.geojson was derived from the data in desire-lines-few.geojson, wondering if there's a bug in the code that generates those percentages.

Or do the percent columns start from a different input dataset, not desire-lines-few.geojson?

[Robinlovelace]

Another question, if the issue is about the percent_ columns and not negative walking values, should the title of this issue be updated accordingly?

[Robinlovelace]

I've found the offending line:

"percent_drive_convertable": -8.0

I don't think this is a priority, worth getting to the bottom of it after the workshop I think.

[Robinlovelace]

Heads-up @joeytalbot I've just created a new milestone for tracking fairly minor issues like this that we can fix after the workshop: https://github.com/cyipt/actdev/milestone/4

You up for taking a look at the issue at some point? Please assign yourself if so. Otherwise I'm happy to take a look and make further updates to the scenario generation code.

[Robinlovelace]

Is this fixed now? I haven't made any major changes, just fixed Bath, so guess not...

[Robinlovelace]

In case it helps @joeytalbot, I now see this, strange to have negative convertible numbers:

all_sites %>% filter(site_name == "tresham") %>% 
+   select(percent_commute_active_base:percent_drive_convertable)
Simple feature collection with 1 feature and 2 fields
geometry type:  POLYGON
dimension:      XY
bbox:           xmin: -0.600986 ymin: 52.49511 xmax: -0.575237 ymax: 52.51319
geographic CRS: WGS 84
# A tibble: 1 x 3
  percent_commute_active_base percent_drive_convertable                                                                                     geometry
                        <dbl>                     <dbl>                                                                                <POLYGON [°]>
1                          12                        -8 ((-0.583048 52.49511, -0.575237 52.50044, -0.581245 52.50327, -0.582275 52.50912, -0.5813...

[joeytalbot]

No it's not fixed.

[Robinlovelace]

No it's not fixed.

You up for taking a look to diagnose the source of the issue? The desire line scenarios looked OK to me when I checked them.

[joeytalbot]

We agreed before that it wasn't a top priority. I'll have a look when I get a chance.