MUMBAI, India, April 17 -- Intellectual Property India has published a patent application (202641042776 A) filed by G Ashwin Prabhu Dr. S. Ramasubramanian; Mr. Karthik Jeevaraj; Ms. Keemathi Kiran; Ms. Devamithra; Mr. Adhil Anvar; Mr. Vigneshwaran V; Mr. V. Vinoth; and Mr. Mohammed Afthab, Chennai, Tamil Nadu, on April 2, for 'intelligent optimization of aircraft fuel consumption using machine learning.'
Inventor(s) include Dr. S. Ramasubramanian; Mr. Karthik Jeevaraj; Ms. Keemathi Kiran; Ms. Devamithra; Mr. Adhil Anvar; Mr. Vigneshwaran V; Mr. V. Vinoth; and Mr. Mohammed Afthab.
The application for the patent was published on April 17, under issue no. 16/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to an Intelligent Machine Learning (ML)-based method and system for optimizing aircraft fuel consumption in real-time and across mission planning phases. The system integrates data from onboard flight management systems (FMS), atmospheric weather sensors, engine performance monitors, and Air Traffic Control (ATC) feeds, enabling comprehensive fuel-efficiency analysis across diverse aircraft types, route categories, and flight phases. The proposed system comprises: (a) a real-time data ingestion module configured to collect multi-source flight parameters at a sampling rate of 10 Hz with latency below 200 ms; (b) a feature engineering pipeline employing multi-variate normalization and dimensionality reduction to derive a unified fuel-state vector; and (c) an ML-based fuel prediction and recommendation model using an ensemble of Gradient Boosted Decision Trees (GBDT) and Long Short-Term Memory (LSTM) networks, trained on over 5 million flight records spanning eight route categories and five aircraft families. The system achieves a fuel consumption prediction accuracy of 97.3%, mean absolute error (MAE) below 1.8%, and recommendation computation time under 0.8 seconds per flight segment, outperforming conventional rule-based fuel management by 38%. Additionally, the ML model delivers adaptive speed, altitude, and routing advisories with a demonstrated fuel savings coefficient of 4.2-7.6% per flight validated against actual airline operational data. The invention further incorporates a continuous learning module that retrains the ML model based on new flight data, improving prediction accuracy over time without manual reconfiguration. The system can be implemented through an embedded avionics processor or cloud-based ground operations interface, making it suitable for commercial aviation, cargo operations, and military transport applications. By combining multi-source flight telemetry with ML-driven optimization, the invention offers a rapid, reliable, and cost-effective solution for sustainable fuel management, operational efficiency, and emissions reduction, compliant with ICAO Annex 6, DGCA CAR Section 8, and IS 15701:2021 aviation standards."
Disclaimer: Curated by HT Syndication.
