The challenge of synergistic optimization for timeliness and cost-effectiveness in transporting high-value goods was addressed. A dynamic route optimization method for freight air-rail intermodal transport was proposed based on a multi-layer temporal network. A multi-objective optimization model integrating strict time windows was constructed by incorporating dynamic schedules from both air and high-speed rail networks. A simulated annealing algorithm with a compound coding strategy was designed to minimize total transport time and cost. The model was validated using a numerical example of 50 freight tasks across 39 hub cities. Results indicate that the proposed method effectively captures the dynamic impact of schedule fluctuations on network connectivity and supports dynamic routing decisions. Comparative analysis shows that the air-rail intermodal mode achieves the highest task completion rate. It reduces the average cost by 38.05% compared to air transport alone and shortens the average time by 19.77% compared to high-speed rail transport alone, achieving a balance between efficiency and economy. Furthermore, the model can flexibly adapt to the transport needs of different cargo types through preference weight coefficients.