Port Weather Navigator

Port Weather Navigator is a web application designed to provide insights into port congestion, predict potential shipping delays due to weather, and offer detailed weather forecasts for various ports.

Features

• Port Selection: Easily select and view data for different ports.
• Weather Panel: Display current and forecasted weather conditions for the selected port.
• Delay Prediction: Predict potential shipping delays based on weather forecasts and historical data.
• Congestion Analytics: Analyze historical and predicted port congestion levels.
• Time Series Forecasting: View time series forecasts for weather and shipping metrics.

Screenshots

Project Structure

The project is divided into two main components:

1. Frontend (port-weather-navigator): A React-based single-page application built with Vite and TypeScript, providing the user interface.
2. Backend (pbl2-ship-delay): A Python-based server that serves data, performs predictions, and exposes API endpoints consumed by the frontend.

Prerequisites

Frontend:

• Node.js (v18 or later recommended)
• npm (or yarn/pnpm)

Backend:

• Python (3.10 or later recommended)
• pip

Cloning and Running the Project

1. Clone the repository:

   git clone https://github.com/MatricalDefunkt/port-weather-navigator
   git clone https://github.com/MatricalDefunkt/pbl2-ship-delay

2. Setup and run the Backend (pbl2-ship-delay):

   cd pbl2-ship-delay
   pip install -r requirements.txt
   python app.py

   The backend server will typically start on a local port http://localhost:5000.

3. Setup and run the Frontend (port-weather-navigator): Open a new terminal window.

   cd port-weather-navigator
   npm install # or yarn install / pnpm install
   npm run dev

   The frontend development server will typically start on http://localhost:5173 and will proxy API requests to the backend.

Technologies Used

Frontend (port-weather-navigator):

• React: A JavaScript library for building user interfaces.
• Vite: A fast build tool and development server for modern web projects.
• TypeScript: A typed superset of JavaScript.
• Tailwind CSS: A utility-first CSS framework.
• Shadcn/ui: Re-usable components built using Radix UI and Tailwind CSS.
• Recharts: A composable charting library.
• TanStack Query: For data fetching, caching, and state management.
• React Router DOM: For client-side routing.

Backend (pbl2-ship-delay):

• Python: Main programming language.
• Flask: Web framework for building APIs.
• Pandas & NumPy: For data manipulation and numerical operations.
• Scikit-learn: For machine learning tasks and preprocessing.
• Keras / TensorFlow: For deep learning models (e.g., RNN, CNN, MLP).
• Joblib & Pickle: For saving and loading machine learning models.

API Endpoints

The frontend communicates with the backend via a set of API endpoints. The main Flask application (app.py) in the pbl2-ship-delay directory defines these routes.

• GET /: Health check for the prediction server.
  • Returns: JSON with status, message, and list of available models.
• POST /predict: Endpoint to make delay predictions.
  • Expects: JSON payload with vessel_type, teu, arrival_timestamp_str, hourly_weather_forecast, and model_name.
  • Returns: JSON with prediction, model used, input summary, and processing time.
• GET /weather: Fetches weather data from Open-Meteo API (with caching).
  • Query Parameters: lat (latitude), lon (longitude), forecast (boolean, default true).
  • Returns: JSON with weather data (current or forecast).
• GET /weather/cache/stats: Retrieves statistics about the weather cache.
  • Returns: JSON with cache stats and configuration.
• POST /forecast/arima/train: Trains a new ARIMA model.
  • Expects: JSON payload with time_series_data (list of values), order (e.g., [1,1,1]), train_test_split.
  • Returns: JSON with training results (RMSE, MAE, etc.) or error.
• POST /forecast/sarima/train: Trains a new SARIMA (Seasonal ARIMA) model.
  • Expects: JSON payload with time_series_data, order, seasonal_order (e.g., [1,1,1,7]), train_test_split.
  • Returns: JSON with training results or error.
• POST /forecast/arima/predict: Makes predictions using a trained ARIMA model.
  • Expects: JSON payload with steps (number of future steps to predict).
  • Returns: JSON with forecasted values or error.
• POST /forecast/sarima/predict: Makes predictions using a trained SARIMA model.
  • Expects: JSON payload with steps.
  • Returns: JSON with forecasted values or error.
• GET /forecast/arima/models: Lists available (trained) ARIMA/SARIMA models.
  • Returns: JSON with a list of model names and their status.