Outbound Path Optimization for Sustainable China–Europe Railway Express Operations Considering Cargo Consolidation and Departure-Time Matching

In outbound China–Europe Railway Express (CR Express) operations, dispersed cargo access, transshipment-node activation, order consolidation path assignment, block train line selection, and departure-time matching are strongly coupled, directly affecting the economic performance, timeliness, and block train operation efficiency of transport plans. Existing studies have mainly focused on the local optimization of consolidation paths, train routes, or departure times. At the same time, insufficient attention has been paid to the bilevel combinatorial structure between node activation and order path assignment. To address this issue, this study proposes an outbound path optimization model for the CR Express considering cargo consolidation, and develops a graph-guided adaptive large neighborhood search (Gra-ALNS) algorithm. The algorithm is organized around a bilevel search structure of node activation and order path assignment. At the upper level, a transport organization graph is constructed, and graph convolutional network (GCN) embeddings are integrated with cargo-access reachability to generate H-node scores, which guide the search for promising transshipment-node activation combinations. At the lower level, outbound generalized paths are used as the basic search units, and upper confidence bound (UCB)-based operator selection, local exact refinement, and cross-ε-grid elite-solution warm starts are combined to optimize order consolidation paths, block train line selection, and departure-time matching. Numerical results show that the proposed Gra-ALNS outperforms conventional adaptive large neighborhood search (ALNS) in computational efficiency and multi-objective solution-set quality. Activating transshipment nodes can reduce the transport cost and in-transit time cost of some orders, but may increase fixed train operating cost and reduce average load factors. The findings provide methodological support for outbound cargo organization optimization, transshipment-node configuration, and block train operation planning of the CR Express. By improving cargo consolidation efficiency, coordinating departure-time matching, and revealing the trade-off between temporal accessibility and train-loading efficiency, this study provides a decision-support reference for a more efficient and sustainable CR Express freight organization.