This paper presents a methodological framework for assessing transfer robustness in multimodal transport networks through graph database technology. Leveraging Neo4j and its Graph Data Science library, a synthetic multimodal network was constructed, comprising diverse node types that represent real-world transport entities, such as bus stops, train stations, metro stations, bike-sharing points, and intermodal transfer hubs. Directed, weighted relationships between specific nodes capture intermodal connectivity and estimated travel times, enabling the application of graph algorithms for evaluating network performance. To measure robustness under disruption, the study introduces the Transfer Robustness Index (TRI), a novel, yet practical and comprehensive metric designed to evaluate intermodal resilience that jointly considers reachability, travel time, and transfer complexity, thus capturing both connectivity preservation and degradation in transfer efficiency. Three TRI variants are considered: binary (reachability), weighted (travel-time efficiency), and adjusted weighted (incorporating transfer complexity via hop count). A scenario-based experiment examines the removal of a critical transfer hub and its impact on multiple origin–destination pairs. Findings indicate that while partial connectivity is preserved, overall intermodal efficiency and simplicity decline substantially. The proposed framework provides a reproducible and adaptable approach for robustness assessment, offering valuable insights for the design and planning of resilient transport systems.