Lightning Risk Prediction for Aviation Safety

Flight tracking systems play an important role in boosting safety, increasing operational efficiency, and keeping an eye on air traffic as it happens. This study introduces a live flight tracking system that gathers aircraft information in real-time through the OpenSky Network Application Programming Interface (API) and can easily grow to handle more data. The system includes a risk analysis module that evaluates possible dangers, like the risk of lightning, by using a Hybrid Temporal Risk Prediction Model (HTRPM). The HTRPM uses Bi-Directional Long Short-Term Memory (BiLSTM) to analyze weather data over time and employs Gaussian Process Regression (GPR) to measure uncertainty. Machine learning, which uses Proximal Policy Optimization (PPO) to improve flight paths, boosts risk predictions by looking at past and current weather data. The system lets users search for flights by choosing their departure and destination airports. A map created with Leaflet.js enhances understanding of the situation by showing where flights are located and indicating the safest paths. With an accuracy rate above 93%, the system shows strong performance in real-life situations. The suggested system provides a complete, real-time way to track flights and reduce risks, helping airline operators, passengers, and air traffic controllers.