NetworkX--Flight-Network-Resilience-using-RAG

We model flight networks as graphs to pinpoint critical vulnerabilities and quantify resilience, then expose these insights through a RAG-powered natural-language interface.

Flight Network Resilience Dashboard

An interactive dashboard to explore the resilience of global flight networks using graph‐theoretic methods and a Retrieval-Augmented Generation (RAG) interface for natural-language queries.

---

Features

• Network Visualization
  Plot the full undirected graph of airports (nodes) and direct flights (edges).

• Centrality Analysis
  Compute and display top-5 airports by degree and betweenness centrality.

• Disruption Simulation
  Compare how the largest connected component (LCC) degrades under:

  • Random failures
  • Targeted attacks on high-centrality nodes

• Shortest-Path Examples
  Show how rerouting changes when a hub (e.g. ATL) is removed.

• Extended Risk Analyses

  • Cancellation Risk (β-distributed edge cancel_rate)
  • Delay-Weighted Path (normal-distributed edge avg_delay)
  • Synthetic Weather-Risk Ranking (β-distributed per-node scores)

• RAG Q&A Interface
  Ask questions in plain English—e.g.

  “Which airport removal fragments the network most?”
  Answers are generated by retrieving key analytics and passing them to an LLM.

---

Requirements

• Python 3.8+
• Streamlit
• pandas
• networkx
• matplotlib
• langchain-openai (optional, for RAG)
• faiss-cpu (optional, for RAG)

1. Clone the Repository

git clone https://github.com/Kusumareddy28/NetworkX--Flight-Network-Resilience-using-RAG.git

2. Install Dependencies

pip install streamlit pandas networkx matplotlib langchain-openai faiss-cpu

3. Run the Application

streamlit run streamlit_app.py

Screenshots

4. Future Enhancements

• Real‐World Weather & Delay Data
  Integrate historical meteorological and flight-delay datasets (e.g., NOAA, Bureau of Transportation Statistics) to replace synthetic risk scores with empirical measures.

• Weighted & Directed Graph Support
  Model edge weights by flight frequency, passenger volume or capacity, and support directed edges for origin-destination asymmetry.

• Temporal Analysis & Animation
  Enable time-sliced resilience (peak/off-peak, seasonality) and animate graph evolution under rolling disruptions.

• Geospatial Map Visualization
  Overlay nodes and edges on an interactive world map (e.g., via PyDeck or Folium) to contextualize vulnerabilities geographically.

• User Authentication & Profiles
  Add login with OAuth/GitHub to save custom scenarios, bookmarks and export personalized reports.

Contributing

If you'd like to contribute to this project:

1. Fork the repository.
2. Create a new branch for your feature or bug fix.
3. Submit a pull request.

Contact

For questions or feedback, feel free to reach out:

Author: Kusuma Reddyvari
Email: kusumareddy28@gmail.com
GitHub: KusumaReddyvari