Roads to Rome

During the last couple of months we worked on solving one of the biggest mobility quests: Do all roads lead to rome? -> roadstorome.moovellab.com

All maps were made with great tools & open data! graph hopper, osm, mapbox, morganherlocker, nodejs, leaflet, mongodb

Roads to Rome

So we have our destination: Rome. But where do you start to find all roads to Rome? For this we created a square grid that covers all of Europe. Every cell of the grid contains the starting point of one route. Next step was combining all 500k routes on one map. For this we aggregated the number of times every street segment was used in the routes. Following this count we created this map:

Roads to US Romes

Continuing on the thought of finding all ways to Rome we set off to the USA: We actually found 10 cities named Rome. This posts a new problem: we don’t have one destination but 10. Starting points are still spread in a grid covering the US. So from every starting point, we searched for the closest of the US Romes. That solves that. Now we color keyed every rome and out came a map, that shows the closest rome to every location in the US:

we continued on the territorial thought, check out more maps here: (roadstorome.moovellab.com)

Urban Mobility Fingerprint / Street DNA

Since we covered the large scale, we continued on a small scale city level by introducing the Mobility Fingerprint and DNA

Creating a N-to-1 Route also works for the city. In our visualisations we created an isochrone of 15min keeping the routes displayed from all segments calculated.

1. Routes are coloured according to their compass direction. 2. Routes are rotated so that they align north to south. 3. Routes are scaled to match start and destination points

Basically the DNA shows the deviation from the ideal path from A to B aggregated with all routes.

I’m really looking forward to your comments. Check our the whole project here: roadstorome.moovellab.com