Skip to content

tosoham/flight-insight

Repository files navigation

✈️ Intelligent Flight Delay Prediction System

📌 Overview

This project is an AI-powered Flight Delay Prediction System that predicts the expected delay duration of flights using machine learning regression models.

The system enables airlines to:

  • Estimate potential delays in advance
  • Optimize flight scheduling and reduce cancellations
  • Improve passenger satisfaction with proactive communication & rebooking strategies

🚀 Features

  • Regression-based Predictions for flight delay duration
  • Data-driven Modeling with historical flight, weather & operational data
  • API Integration using FastAPI for predictions
  • Frontend Dashboard (React + Tailwind) for users & airline staff
  • Dockerized Microservices for easy deployment

🏗️ System Architecture

Data Sources

  • Historical flight datasets
  • Real-time weather feeds
  • Operational & airport-level information

Processing Pipeline

  • Data cleaning & preprocessing
  • Feature engineering (weather, congestion, seasonal patterns, etc.)
  • Regression model training & validation

Prediction Layer

  • FastAPI-based ML microservice (ml_service)
  • Predicts expected delay (in minutes)
  • Provides APIs for integration

Frontend

  • Built with React + Tailwind CSS
  • Displays predictions, insights, and dashboards

Deployment

  • Docker & Docker Compose for containerization
  • Optionally extendable to Kubernetes

📊 Business Impact

  • Reduce flight disruptions & cancellations
  • Improve customer satisfaction with accurate delay forecasts
  • Optimize airline resources → higher ROI

🛠️ Tech Stack

  • Languages: Python, SQL, JavaScript (React)
  • ML Libraries: Scikit-learn, XGBoost/LightGBM, Pandas, NumPy
  • Backend (ML Service): FastAPI
  • Frontend: React + Tailwind CSS
  • Deployment: Docker, Docker Compose
  • Database (optional): PostgreSQL / SQLite

📂 Repository Structure

📦 flight-insight
 ┣ 📂 backend_dev         # (optional backend experiments / services)
 ┣ 📂 ml_service          # Machine Learning microservice
 ┃ ┣ 📂 app               # FastAPI app
 ┃ ┃ ┣ 📄 main.py         # Main FastAPI entrypoint
 ┃ ┃ ┣ 📄 __init__.py
 ┃ ┣ 📂 databases         # DB-related configs (if used)
 ┃ ┣ 📂 model             # ML models / training scripts
 ┃ ┣ 📄 requirements.txt  # Python dependencies
 ┃ ┣ 📄 Dockerfile        # Dockerfile for ML service
 ┃ ┣ 📄 alembic.ini       # DB migrations (if using Alembic)
 ┃ ┗ 📄 init_db.py        # DB initialization script
 ┣ 📂 frontend            # React frontend
 ┃ ┣ 📂 src               # Components, pages, hooks
 ┃ ┣ 📂 public            # Static assets
 ┃ ┣ 📄 package.json      # Frontend dependencies
 ┃ ┣ 📄 Dockerfile        # Dockerfile for frontend
 ┣ 📂 docs                # Documentation files
 ┣ 📄 .env.example        # Example environment variables
 ┣ 📄 docker-compose.yml  # (optional) service orchestration
 ┣ 📄 README.md           # Project documentation
 ┗ 📄 LICENSE             # License information



---

🤝 Contributing

We welcome contributions! Please fork this repo, create a feature branch, and submit a pull request.


📧 Contact

For queries, collaborations, or feedback:
Team CTS NPN Hackathon Project – [Your contact info or team email]