unbrother / traffic-major-minor Goto Github PK

I have managed to create some of the variables except for the aadt of closest major road, mentioned in the original methodology. I know there this would be something similar to what was done in the original code shared with me by @mem48 but can't find the correct steps to do it.
@Robinlovelace

The original model based the classification of roads as urban or rural from the settlement Ordnance Survey Meridian 2, the paper does not mention how (I assume it can be reverse-engineered from the .sql files), but I am not sure on how to do this with R. I can't find the Meridian 2 files online but from the Strategi files, I found there's a layer named urban_region, which I could intersect with roads. I know this process would be very similar to the one that assigns AADT to roads but I'm unsure on how to do it.

@mem48
@Robinlovelace

In osm the maxspeed tag is in kph so 60 means 60 kph while 60 mph means 60 mph. But in the UK people sometime forget this and miss the mph.

Two issues from this:

1. You should clean the tags by adding mph so that dodgr treats them correctly

2. In the UK roads are 60 mph unless otherwise specified. This means that there are narrow windy country roads that are 60 mph legally but are practically more like 30 - 40 mph. If you can think of an easy way to detect these and lower the maxspeed it might improve results. But this is low priority at the moment.

The road importance part of the postGRE code first makes subdivisions enclosed by the major roads. I am not sure how to do this in R, I believe I would just need to draw a polygon between those lines but do not know how to make R select the enclosed areas.
After that they made something different for roads outside of these areas, which they name coastal roads. I have not checked this in full but I am sure it will be a challenge for me.

Until now, I used dodgr with weight_streetnet() to get some kind of relative weighting and I was considering that as the road importance. But this is not the process originally used.

@mem48 and @Robinlovelace, do you have any pointers or ideas on how to attack this part?