Urban Road Safety

Last modified on 2022-02-28

This repository follows the paper Identifying urban features for vulnerable road user safety in Europe in which we identified urban features that are determinants of vulnerable road user safety through the analysis of inter-mode collision data across European cities. Our regression results (shown in the heatmap below) suggest that policies aimed at increasing the modal share of walking and cycling are keys to improve road safety for all road users.
Here, we share the extracted/computed urban features and inter-mode collision data, as well as the regression task notebook.

(Transportation icons designed by Freepik.)

---

Requirements

• Python, Jupyter
• Pandas, NumPy, Matplotlib, Seaborn
• StatsModels, Scikit-Learn
• (OSMnx - used to generate the published data, not needed for the reproduction of the regression results)

Data

The CSV data file contains 24 cities in Europe (rows), their respective urban features and computed inter-mode collision data for all crash participant pairs (columns).

Regression Notebook

The notebook performs the regression task on 6 selected crash participant pairs.
Before running the notebook make sure to have all the requirements installed and the data downloaded locally with path adjustment in the code.

Authors

This project was developed by Marina Klanjčić, Laetitia Gauvin, Michele Tizzoni and Michael Szell during Marina's Lagrange applied research fellowship at Data Science for Social Impact and Sustainability Research Group of ISI Foundation in Turin, Italy.

Data and code-related questions should be addressed to Marina Klanjčić (mrnkln@outlook.com).

Acknowledgments

This work was supported by the Lagrange Project of the CRT Foundation.

MIT License

Back To The Top