Inspiration

Scooters can really get in the way of pedestrians. In the future these scooters must only be parked inside designated stations. Question remains where best to place them.

What it does

Using data from multiple sources, we build up a model of where people are and how they will connect to the metro. With VIA-resistance model we can compare resistance of walking directly to S+U Bahn stations to using scooters to S+U Bahn stations. Based on this estimation, we perform simulations to figure out where best to place scooter stations.

How we built it

Explore data we can get our hands on and try to gain insight. Python simplifies this process. We then build everything around this data and code a prototype.

Challenges we ran into

Not so easy to create this from the data we have. We had to check back to statistics we have to check the sanity of our model. Performing full-fledged simulations also not possible without algorithm engineering.

Accomplishments that we're proud of

Despite indirect solutions, we still managed to get a sensible and usable prototype. We also succeeded in learning web scraping in an hour.

What we learned

Web scraping! Calculations in the real world is also very complex and compromises have to be made.

What's next for OptiScooti

More accurate infos on population, for example include demographics (age, job) to gain more accurate models. Other data sources (BVG route plans, recent scooter trips, ...) can also be combined into the model. Better optimization algorithms could also be useful (evolutionary, ...). Most importantly, feedback from actually placing scooters there will provide powerful insights. Since stations will be coming, there's limited time to experiment!

Built With

Share this project:

Updates