Transformer-Augmented MCTS for Aircraft Landing Problem

The Aircraft Landing Problem (ALP) poses significant challenges for traditional Monte Carlo Tree Search (MCTS) due to its vast search space and reliance on inefficient random simulations. To overcome these limitations, this paper proposes a novel Transformer-Augmented Monte Carlo Tree Search (TMCTS) algorithm. Our approach integrates a reinforcement learning framework that incorporates key operational constraints, including wake turbulence separation and time windows, and employs a cost function aimed at minimizing both delay time and fuel consumption. A core innovation is the replacement of the conventional random simulation phase in MCTS with a Transformer-based value predictor. This leverages the Transformer’s superior capability in sequence modeling and capturing global dependencies among flights, thereby dramatically accelerating search convergence. Specifically, we design a two-head Transformer network (comprising policy and value heads) to provide informed prior knowledge, which effectively guides the selection and expansion steps of the MCTS tree. The model is trained within an Actor-Critic framework, utilizing behavior cloning for pre-training followed by reinforcement learning for fine-tuning. Experimental evaluations on the standard OR-Library benchmark demonstrate that our TMCTS method significantly reduces scheduling deviation compared to state-of-the-art baselines (including DPALO+GA, DPALO+PSO, and DALP). Moreover, it achieves a 90.6% reduction in computation time relative to the DALP method, highlighting its superior efficiency and practical applicability for real-time scheduling.