Sustainable Travel Planner

prototype.mov

Sustainable Travel Planner helps users compare train and plane carbon emissions and travel duration. The application is built using Streamlit, map and charts are created in Vega-Altair.

Three versions of the sustainable travel planner with different data visualisations were created for user testing:

• Version 1: horizontal bar charts
• Version 2: vertical bar charts
• Version 3: bullet chart and proportion area circles

This tool is part of a Cartography MSc. thesis "Maps for sustainable business travel advice" (Poster, Presentation), conducted by Viktor Kochkin at the ITC, University of Twente.

Features

• Compare carbon emissions and travel duration between train and plane.
• Visualize travel routes on a map.
• Support for round trips and multiple passengers.
• Dynamic updates based on selected cities.

Project Structure

└── sustainable_travel/
    ├── streamlit_app.py    #Streamlit app main script
    ├── utils.py        #Functions for data loading, normalization, and map creation
    ├── data_gathering.ipynb   #Data preparation scripts in jupyter notebook
    ├── data
    │   ├── coordinates.csv #Coordinates of 29 cities for the map
    │   └── trips_data.csv  #Plane and train travel time and emissions data for 406 city pairs
    ├── geojson_files
    │   ├── lines  #Train routes polylines
    │   └── points #Train routes transfer points
    └── pages # Different prototype versions for user testing
        ├── ver1.py
        ├── ver2.py
        └── ver3.py

Scope and Limitations

Route planning is limited to 29 European cities:

- Amsterdam
- Bern
- Bilbao
- Bratislava
- Brussels
- Bucharest
- Budapest
- Copenhagen
- Dresden
- Istanbul
- Lisbon
- Ljubljana
- London
- Luxembourg City
- Madrid
- Munich
- Oslo
- Paris
- Prague
- Riga
- Rome
- Sofia
- Stockholm
- Tallinn
- Vienna
- Vilnius
- Warsaw
- Zagreb

Data Flow

1. Loading Data:

   • The trip and coordinates data are loaded from CSV files into Pandas DataFrames.
   • Column names are stripped of any leading or trailing spaces.

2. Normalizing Data:

   • City pairs in the trip data are normalized using the normalize_city_pair function.

3. Creating Maps:

   • The base map is created using the create_base_map function, which includes city points.
   • GeoJSON routes are loaded using the load_geojson_route function to draw routes on the map.

4. User Interaction:

   • Users select departure and destination cities, number of people, and whether the trip is a round trip.
   • Travel details are displayed, including duration and CO2 emissions for both train and plane.
   • Dynamic charts are created using Vega-Altair to visualize travel duration and CO2 emissions.

5. Dynamic Updates:

   • The app dynamically updates the 'To' city options based on the selected 'From' city.
   • Travel details and maps are updated based on user input.