Futureflow

Inspiration

As cities grow, the public transport system seems to never be able to truly catch up, endless delays, longer travel time due to too many privately owned cars and overcrowded public transport means. What if this could be solved by using a new dynamic product that allows public and private companies to adapt to the growth and the dynamic events happening within the city?

What it does

Futureflow is a dashboard that fetches data from publicly available sources in real time (e.g. APIs, event websites, government transportation databases...), organizes it into different categories, interprets it in order to show and predict how crowded public transport is within specific city parts. This framework can then be provided to public and private transport companies in order to improve the transportation system within a city by making it more efficient

How we built it

We used the scrapy framework in order to get data from the internet, once we got the data we classified it into different groups and stored it in a database. Once the data is saved we utilize several python programs that use the event data to generate load numbers within the specific city parts at different times. Afterwards we feed the data to the API endpoints and the react framework makes requests to the API. Later on it creates a dashboard where on the left side it creates 2 tabs, one where the load for each station within the city can be read and another tab where the future forecasts of events can be read. On the right side it shows openstreetmap with dots that change color depending on the amount of load at each city part in comparison to the normal load amount.

Challenges we ran into

Classifying and gathering data from the different events happening throughout the city and use the data with our framework, as well as making the map update in real time.

What we learned

We learned how to work with and get geological data from different sources, as well as classifying event data.

What's next for Futureflow

We will provide a new framework for governments and private companies to use in order to determine and react upon specific traffic loads within the city and make the public transport data more accurate, therefore making public transport more efficient. Improving the quality and quantity of data we get would be one of our priorities, as well as categorizing the data in a more concise manner. An important addition we would incorporate is a feedback element into the system. This would enable different companies to provide their load data, which, through machine learning, would allow for continuous improvement of the system.