Open gasato opened 3 years ago
Hi @gasato thanks for the message and sorry for the belated response. Short answer: I think so! In fact @dabreegster has developed an alternative data driven way of dividing up urban areas into zones using the road network.
Hi Gabriel,
If I understand your idea correctly, you're interested in how network connectivity is between adjacent zones: The example blue lines could be major roads or transit routes. In some cities, maybe there are more options to go between zones close to the center, but it's more bottlenecked farther out. Is that what you mean?
I'm working currently on something a bit different -- a way to physically partition 2D space based on the edges of roads. https://a-b-street.github.io/docs/software/ltn/tech_details.html#neighborhood-selection has a very brief explanation, and you could get a quick intuition for the work at http://play.abstreet.org/0.3.9/ltn.html?system/us/seattle/maps/phinney.bin by clicking a neighborhood, then doing "adjust boundary" to expand the selection block-by-block. I'm using this to study low-traffic neighborhoods, and a first question is exactly how to define the boundary of a neighborhood -- do natural or man-made features like rivers, parks, and railroads split things? I'd be happy to talk more about this work if you're interested; I'm also working on a paper.
Hi Rovin First of all congratulations for your presentation at FOSS4G in Buenos Aires.
One idea: Have any sense to have the possibility to transform the zone center to a zone network diagram with all their neighborhood connections?
Let me explain. From the center to the outer sectors of your diagram you have a connectivity relationship. Same with adjacent sectors of each sector. Do you think it will be worth to model your result not only as a zones but also a network? I think networking modelling can also allow to make studies about the influence of the center as you move away from it. I am proposing to have the posibility to translate your zone result to a networks diagram too in order to study connectivity phenomenas.
Regards Gabriel