Nextbike City Analysis

Analyze NextBike trips in your city: collect, process, and visualize bike trips.

🔴 Live Preview

Overview

- collection/       # Setup data collection
- data/             # Stored trips data
- docs/             # Documentation
- processing/       # Calculate trips
- visualization/    # Web visualization

Prerequisites

• Docker with Compose (Install Docker)

🚀 Install

1. Clone the repository:

   git clone https://github.com/zwoefler/nextbike-city-analysis.git
   cd nextbike-city-analysis

2. Create your .env file:

   cp .env.example .env

   Edit .env and set your values:

   Variable	Description
   DB_USER	Postgres username
   DB_PASSWORD	Postgres password
   DB_NAME	Postgres database name
   DB_HOST	Postgres host (nextbike_postgres when using compose)
   DB_PORT	Postgres port (default: 5432)
   DB_BIKES_TABLE	Table name for raw bike data
   DB_STATIONS_TABLE	Table name for station data
   DB_CITIES_TABLE	Table name for city data
   CITY_IDS	Comma-separated Nextbike city IDs to collect, e.g. 467,210
   STATIONS_SYNC_INTERVAL_HOURS	How often to sync stations
   CITIES_SYNC_INTERVAL_HOURS	How often to sync city metadata
   EXPORT_DIR	Output folder for processed trip files (default: /data)
   VISUALIZATION_PORT	Port for the web UI (default: 8080)

   Find your city ID in city_ids_2025_02_15.md.

3. Start everything:

   docker compose up -d

   This starts four services:

   • postgres - database for raw bike, station, route, and trip data
   • collector - polls the Nextbike API every 60 seconds and writes to postgres
   • processor - runs at midnight; calculates trips for each city in CITY_IDS, caches OSM routes in the database, and writes .geojson.gz files to the trip_data volume
   • visualization - FastAPI app at http://localhost:${VISUALIZATION_PORT} that serves trip and station data live from the database

4. Open http://localhost:8080 (or your configured VISUALIZATION_PORT) in your browser.

   ⚠ To trigger processing manually, see docs/manual-processing.md.

Stop / destroy

# Stop containers
docker compose down

# Stop and delete all data (including the database volume)
docker compose down -v --remove-orphans

Credits

Inspired by 36c3 - Verkehrswende selber hacken by ubahnverleih & robbie5.

Visualization inspired by Technologiestiftung Berlin:

• Bike-Sharing
• Bikesharing-Vis