Traffic Detection and Optimization Project

This repository contains a comprehensive traffic detection and optimization project for Berlin. It integrates traffic, weather, public transportation, and construction data to build predictive models and optimize routes.

Data Sources

• Traffic Data:

  • Source: Berlin Municipality Traffic Management Division.
  • Access: Berlin's official website provides traffic flow and congestion data. The Traffic Detection Berlin API is also utilized.
  • Dataset: Traffic Detection API

• Weather Data:

  • Sources:
    • Open-Meteo: High-resolution free weather data.
    • WeatherAPI.com: Real-time and historical weather data.
    • OpenWeatherMap: Provides diverse weather APIs.
  • Dataset: Open-Meteo JSON API.

• Public Transport Data:

  • Source: Berliner Verkehrsbetriebe (BVG).
  • Access: Developer portal and Trafi platform for integrated public transportation schedules and routes.
  • Dataset: BVG GTFS Data

• Road and Construction Information:

  • Source: Berlin Municipality's announcements on roadworks and construction projects.
  • Access: Official website provides updates on roadworks and restrictions.
  • Dataset: Roadwork Data API

Code Overview

Installing Required Libraries

To set up the environment, install the required Python libraries:

pip install -r requirements.txt

Key Modules

• data_processing.py: Functions to load, clean, and preprocess datasets.
• modeling.py: Machine learning models for traffic prediction.
• visualization.py: Visualization tools for spatial and temporal traffic patterns.

Highlights

• Predictive Modeling:

  • Linear Regression
  • Random Forest
  • K-Means Clustering
  • Gradient Boosting (XGBoost, CatBoost)
  • Neural Networks

• Route Optimization:

  • Integrates OpenRouteService API for route calculations and optimizations.
  • Visualizes road networks and routes using OSMnx.

• Data Integration:

  • Combines traffic, weather, construction, and public transportation data.
  • Enhances features with additional metrics like holiday indicators and weekend information.

Visualization Examples

Traffic Clustering

Using K-Means clustering to identify traffic patterns:

Road Network Visualization

Berlin road network with optimized routes:

Model Comparison

Performance comparison of models based on MSE and R² scores:

How to Use

1. Clone this repository:

git clone https://github.com/busrayatlav/traffic_detection.git

2. Navigate to the project directory:

cd traffic_detection

3. Install dependencies:

pip install -r requirements.txt

4. Run the Streamlit app locally:

streamlit run app.py

License

This project is licensed under the MIT License.

Acknowledgments

Special thanks to the Berlin Municipality and data providers like Open-Meteo, BVG, and OpenRouteService for making this project possible.