Introduction

As a frequent commuter in Paris, the daily journey can sometimes be a challenge. The bustling metro, while efficient, can become overwhelming during peak hours. Conversely, cycling offers an alternative that promises fresh air and freedom but raises questions about efficiency. This project aims to provide a clear comparison between cycling and metro commute times to the Notre-Dame Cathedral from various parts of Paris. The goal is to help commuters make informed decisions about their daily travel.

Motivation

The motivation behind this project stems from personal experience. Navigating the Parisian streets and the metro system daily, I often wondered whether cycling would be a faster and more pleasant alternative. This curiosity drove the creation of a comprehensive visualization that could answer this question not just for me but for all Parisians.

Data Sources

To create an accurate and informative comparison, the project relies on multiple data sources:

Google Maps API:
- Used to fetch cycling times from various points in Paris to the Notre-Dame Cathedral.
- API Key: Provided for access.
Île-de-France Mobilités Open Data:
- Accessed GTFS (General Transit Feed Specification) data for public transit schedules and routes.
- Data includes walking, waiting, riding, and transferring times.
OpenStreetMap (OSM):
- Provided the geographic data required for the visualization.

Tools and Technologies

Python: The primary programming language used for data processing and visualization.
R5py: A Python library designed for transportation analysis, built on top of the open-source R5 routing engine by Conveyal.
Geopandas: Used for geospatial data manipulation and analysis.
Folium: Utilized for creating interactive maps.
BeautifulSoup: A Python library for parsing HTML and extracting data.

Methodology

Step 1: Data Collection

Google Maps API: Retrieved cycling times for a grid of points evenly spaced throughout Paris.
GTFS Data: Downloaded from Île-de-France Mobilités Open Data to compute metro travel times.

Step 2: Data Processing

Hexagonal Grid Creation:
- Created a regular hexagonal grid of points to cover the city of Paris.
- Ensured comprehensive spatial coverage to provide detailed insights.
Travel Time Calculation:
- Cycling Times: Calculated based on distance, assuming a constant speed of 15 km/h.
- Metro Times: Computed using R5py, accounting for the entire trip duration (walking to the stop, waiting, riding, transferring, and walking to the destination).

Step 3: Visualization

Data Integration:
- Combined cycling and metro times into a single GeoDataFrame.
- Merged based on the centroids of the hexagonal grid cells.
Comparison Analysis:
- Determined the faster mode of transportation for each hexagonal cell.
- Used a color-coded system to represent areas where cycling is faster (shades of orange) versus areas where the metro is faster (shades of purple).
Interactive Map Creation:
- Utilized Folium to create an interactive map with tooltips and a legend.
- Provided a user-friendly interface to explore the data.

Results

The final output is an interactive map that clearly shows which areas of Paris are better served by cycling versus the metro. The map is color-coded:

Orange Shades: Faster by bike.
Purple Shades: Faster by metro.
Grey: No data available.

Conclusion

This project offers valuable insights into the efficiency of cycling versus metro travel in Paris. It serves as a practical tool for commuters to optimize their daily travel, potentially saving time and enhancing the commuting experience. By leveraging open data and modern geospatial analysis techniques, this project contributes to a more informed and connected urban mobility landscape.

Acknowledgments

Google Maps: For providing API access to cycling times data.
Île-de-France Mobilités: For making GTFS data publicly available.
OpenStreetMap: For providing geographic data.
Conveyal: For developing the R5 routing engine and R5py library.

Paris Commute Comparison - Bike vs. Metro