search

ATFutures / who

Active Transport Futures planning tools for the World Health Organisation
8 stars 2 forks source link

Implement distance decay with popdens origins #3

Closed mpadge closed 6 years ago

mpadge commented 6 years ago

Use trip origins from #2; allocate trip destinations randomly but according to defined trip distance decay function

mpadge commented 6 years ago

Largely done, but raises one important question in regard to the maps shown here. Using origins defined from population density layer and propagating to all possible destinations according to the logistic polynomial distance decay function means that the (network) locations of peak flow become spatially disconnected from the points of origin. They are effectively radiated randomly outwards to around 3-4km away in network terms, and that radiation with the from-to-everywhere routing becomes effectively a diffusion process. I suspect this needs to be thought about a bit more ...

mpadge commented 6 years ago

solved via dodgr_spatial_interaction which explicit implements an exponential interaction model, and is now incorporated into this script.