Combining GTFS and OSM Data to Identify Critical Infrastructure

This project analyzes Krakow's transport network by combining:

• GTFS data: Public transport schedules and routes
• OSM data: Street network infrastructure

The goal is to identify critical nodes and segments using graph centrality metrics and to visualize the integrated network.

Project Report

Interactive Maps & Plots

• GTFS Data Loading & Preprocessing: Reads and prepares public transport data.
• OSM Data Integration: Downloads and processes Krakow’s street network.
• Segment Utilization Analysis: Calculates and visualizes the busiest and least-used segments.
• Network Graph Construction: Builds a stop-to-stop directed graph and computes centrality measures.
• Multimodal Network Integration: Merges public transport and road networks for holistic analysis.
• Centrality Analysis: Computes eigenvector, PageRank, betweenness, and degree centralities for nodes and edges.
• Critical Node Identification: Highlights the most important nodes for network resilience.
• Interactive Visualizations: Generates interactive maps and charts (HTML output) for all key analyses.

.
├── krakow_transport_analysis.ipynb   # Main analysis notebook
├── gtfs/                            # GTFS data (unzipped)
├── gtfs.zip                         # GTFS data (zipped, optional)
├── results/
│   ├── gtfs_graph/                  # GTFS-based analysis outputs
│   ├── osm_graph/                   # OSM-based analysis outputs
│   └── merged_gtfs_osm/             # Multimodal network outputs
├── img/                             # Images and figures
└── README.md                        # This file

• Python 3.8+
• Recommended packages:
  • pandas, numpy, networkx, osmnx, geopandas, matplotlib, plotly, folium, shapely, scipy, joblib

Install all dependencies with:

pip install -r requirements.txt

Or manually:

pip install pandas numpy networkx osmnx geopandas matplotlib plotly folium shapely scipy joblib

• Place your GTFS zip file as gtfs.zip in the project root.
• The notebook will extract it automatically to gtfs/ if not already present.

1. Open krakow_transport_analysis.ipynb in Jupyter Notebook or VS Code.
2. Run all cells step by step.
3. Outputs (interactive maps, charts, CSVs) will be saved in the results/ subfolders.

1. Load and Prepare GTFS Data
   Reads GTFS files, preprocesses shapes and trips.

2. Analyze Transit Operations
   Counts trips per route/direction, visualizes busiest lines.

3. Preprocess Route Shapes
   Converts shape points into route segments.

4. Calculate Segment Utilization
   Aggregates trip counts per road segment, identifies top and bottom segments.

5. Build Stop Network Graph
   Constructs a directed graph of stops, computes centrality measures.

6. Integrate OSM Street Data
   Downloads Krakow’s road network, converts to graph.

7. Create Multimodal Network
   Merges GTFS and OSM graphs, assigns flow capacities.

8. Assigning Flow Capacities
   Assigns flow capacities to network edges based on road type and bus frequency.

9. Multimodal Centrality Analysis
   Calculates eigenvector, PageRank, betweenness, and degree centralities for nodes and edges.

10. Critical Node Analysis
    Identifies and visualizes the most important nodes for network resilience.

• Interactive HTML maps: In results/, e.g. 2_trips_per_route_direction.html, 4_5_segment_trip_distribution_interactive.html, etc.
• CSV files: Centrality measures and statistics.
• Plots and charts: For all major steps.

• Some output HTML files may be large (>50MB) and are best viewed locally.
• For large files, consider using Git LFS.
• All code is modular and can be adapted for other cities or datasets.

MIT License

• OSMnx
• GTFS
• Folium
• Plotly

Created as part of a master's project on urban transport network analysis.