Fuel consumption is also one of the most affecting factors in the cost of operation as well as the environmental performance in commercial aviation. In this respect, the heightened regulatory pressure and rising fuel prices have underscored the need to adopt data-driven schemes that can effectively regulate fuel consumption and facilitate sustainable airline operations. This paper presents a detailed data analytics system for sustainability optimisation in airline fuel management, combining operational parameters, environmental conditions, and machine-learned fuel-burn prediction. The proposed model shows high predictive performance based on an aviation-realistic synthetic dataset with an R 2 of 0.93 and a 18% decrease in the mean absolute error compared to using baseline regression methods. Numerical simulation also indicates that predicted fuel burn can be reduced by as much as 12 by optimising cruise altitude and reducing holding time, which is consistent with the current research in sustainable aviation analytics and digital-twin-based fleet monitoring. The framework is implemented to be used in the airline operations centres, trajectory-planning systems, as well as the predictive-maintenance environments to provide real-time decision-support to the fuel-efficient flight operations. On the whole, the findings demonstrate the opportunities of integrated data analytics to improve sustainability performance, lower the operation costs, and facilitate next-generation intelligent aviation systems.