• Optimal Schedule Recovery for the Aircraft Gate Assignment with Constrained Resources
  Asadi, E., Schultz, M., & Fricke, H. (2021). Computers & Industrial Engineering.

• A Review on Airport Gate Assignment Problems: Single versus Multi Objective Approaches
  Daş, G. S., Gzara, F., & Stützle, T. (2020). Omega.

• Flight Gate Scheduling: State-of-the-art and Recent Developments
  Omega (2007).

• An Airport Stand Assignment Problem Considering the Passenger Boarding Distance
  Huyan, Z., Tang, T.-Q., Gao, Y., & Cao, F. (2023). Journal of Advanced Transportation.

• A Novel Deep Reinforcement Learning Approach for Real-Time Gate Assignment
  Li, H., Wu, X., Ribeiro, M., Santos, B. F., & Zheng, P. (2024). SSRN.

• A Two-Stage Optimization Model for Airport Stand Allocation and Ground Support Vehicle Scheduling
  Yao, M., Hu, M., Yin, J., Su, J., & Yin, M. (2024). Applied Sciences.

Title: Gate Assignment Algorithm for Airport Peak Time Based on Reinforcement Learning

• Gate Assignment Algorithm for Airport Peak Time Based on Reinforcement Learning—Chenwei Zhu, Zhenchun Wei, Zengwei Lyu, Xiaohui Yuan, Dawei Hang, Lin Feng, 2024. (n.d.). Retrieved August 18, 2025, from

Approach:

• Proposes a two-stage PPO-based model (GABPPO) with pre-assignment and dynamic reassignment.
• Optimizes near-gate usage and passenger transfer efficiency.

Results:

• Higher near-gate assignment rate (76.6%) and better gate matching than DQN/PG/APGA.

Strengths:

• Joint optimization of airport and passenger objectives.
• Real-time reassignment included.

Limitations:

• Small dataset (30 flights).
• No scalability or runtime analysis.
• Lacks comparison with stronger DRL baselines.

Title: Deep Reinforcement Learning for Real-Time Airport Gate Assignment

• Li, H., Wu, X., Ribeiro, M., Santos, B., & Zheng, P. (2025). Deep reinforcement learning approach for real-time airport gate assignment. Operations Research Perspectives, 14, 100338.

Approach:

• Uses A3C for pre-assignment and a real-time reassignment agent (REGAPS).
• Action masking applied for feasibility filtering.

Results:

• Reduces apron usage and passenger walking distance vs. rule-based baselines.
• Handles delayed flights adaptively.

Strengths:

• Real-world dataset (124 flights).
• Handles dynamic delays and gate-sharing constraints.

Limitations:

• Simplifies gate occupation logic (e.g., dep delay ignored).
• No evaluation against other DRL agents like PPO or DQN.

Title: A Deep Reinforcement Learning Algorithm for AGAP

• Jia, A., Liu, H., Yang, H., Yang, W., Chen, H., & Yi, X. (2026). A Deep Reinforcement Learning Algorithm for Solving Airport Gate Allocation Problem. In M. Mahmud, M. Doborjeh, K. Wong, A. C. S. Leung, Z. Doborjeh, & M. Tanveer (Eds.), Neural Information Processing (pp. 75–88). Springer Nature.

Approach:

• DQN-based assignment with hybrid rewards and conflict masking.
• Evaluated on both synthetic benchmarks and large real-world cases.

Results:

• Solves 244-seat/102-gate cases in <2s.
• Outperforms FBS and RFH in quality and runtime.

Strengths:

• High scalability.
• Considers transfer vs. non-transfer passenger walking distance.

Limitations:

• Focused on static schedules.
• No dynamic reassignment or delay modeling.

• A Comprehensive Survey on Graph Neural Networks
  Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2021). IEEE TNNLS.

• ROBOT LEARNING, Edited by Jonathan H. Connell and Sridhar Mahadevan
  Robotica (1999).

• OpenSky
  OpenSky Network – Real-time air traffic data platform.