Heterogeneous data-driven evaluation method for the transfer efficiency of multimodal comprehensive passenger hubs

With the continuous expansion and increasing complexity of comprehensive transport hubs, the efficiency of multimodal connection and transfer within comprehensive passenger hubs directly affects the quality of hub services and the travel experience of passengers. However, the diverse and dynamically changing spatial–temporal travel demands of passengers pose significant challenges for the design optimization and operational scheduling of multimodal connection networks. Therefore, scientifically evaluating and analyzing the performance of these networks has become a crucial step in improving the efficiency of urban comprehensive transport systems. This study takes major comprehensive passenger hubs in Beijing—such as high-speed railway stations and airports—as research objects and constructs a research framework of “multimodal connection network modeling–performance quantification–mechanism analysis.” By integrating multisource heterogeneous data including arriving passenger flows, metro services, ground buses, taxis, ride-hailing orders, and in-station transfer times, and by identifying inter-layer coupling nodes based on spatial rules, the study establishes a fourdimensional, 14-indicator quantitative evaluation system encompassing transfer efficiency, network centrality, network efficiency, and robustness. The entropy weight method is applied to achieve objective weighting, while the cloud model is used to characterize indicator uncertainty, thereby accomplishing the quantitative evaluation of multimodal connection performance in comprehensive passenger hubs. The results reveal significant spatiotemporal heterogeneity in connection performance: during peak hours, network efficiency and centrality are higher but transfer times fluctuate greatly; during nighttime, network robustness decreases and the overall performance declines. The proposed analytical method for evaluating the multimodal connection network performance of external passenger hubs provides a systematic approach to assessing connection service efficiency and quantifying the spatiotemporal dynamics of hub connectivity. This research offers a scientific basis for improving the efficiency of hub transfer services and optimizing passenger travel experiences, and serves as an important reference for the planning, operational scheduling, and management of hub connection networks.