This research integrates Large Language Models with Reinforcement Learning to create more effective bus holding control strategies, addressing traditional challenges in public transportation management.

• Overcomes limitations in bus state prediction and passenger demand estimation
• Leverages data-driven approaches instead of traditional model-based methods
• Enhances operational stability and efficiency of bus systems
• Demonstrates practical application of AI in critical infrastructure

For transportation engineers, this advancement represents a significant shift toward intelligent transit systems that can adapt to real-time conditions, improving both reliability and resource utilization in public transportation networks.

Large Language Model-Enhanced Reinforcement Learning for Generic Bus Holding Control Strategies