Stochastic Cruise Speed Control for Time-Based Metering Under Uncertainty

With the increasing demand for air transportation, implementing safe and efficient air traffic management strategies, such as time-based metering operations, which manage the calculated time over (CTO) at designated fixes, has become essential. These operations rely on precise speed control to ensure that aircraft meet the required CTOs. However, the existing deterministic approaches for speed control do not consider the inherent uncertainties in aircraft operations, particularly those attributed to weather prediction inaccuracies. This study proposes a Markov decision process framework for time-based metering during cruise to address the critical tradeoff between delay uncertainty and operational costs. The proposed framework incorporates the number of speed changes into the state space, as frequent speed adjustments are operationally undesirable and significantly increase the workload of pilots and air traffic controllers. To mitigate this, the proposed approach accounts for the number of speed changes and optimizes both their speed and timing under uncertainty, thereby minimizing delays and fuel consumption while enhancing operational feasibility. The proposed method is validated through numerical simulations, demonstrating its ability to achieve optimal performance with a limited number of speed changes while ensuring adherence to CTOs. The framework achieves approximately 1–2% fuel savings with fewer speed changes, reducing delays and simplifying operations. This study demonstrates the potential of stochastic cruise speed control to enhance air traffic management by integrating realistic uncertainty models and operational constraints.