<![CDATA[The vulnerability of road networks under disaster conditions is typically discussed in terms of hazard exposure. However, the impact of network structure on the consequence of disruption remains an under-explored area. The present paper examines the effect of road network structure on the degradation of connectivity under disruptions caused by floods and landslides. This is achieved through a graph-based analytical framework, with a representative empirical testbed: the primary road network of Central Java, Indonesia. The network is modeled as a graph and the segments at risk of hazard exposure are identified using spatial overlay. These segments are then systematically deleted based on hazard type, hazard level and spatial localization scenarios. The performance of connectivity is measured based on the largest connected component, the average shortest path length and global efficiency. These measurements are taken both with and without the integration of toll roads. The results obtained demonstrated the presence of disruption signatures, each of which is unique to each hazard. Despite the considerable spatial extent of disruptions, the disruptions caused by flooding result in a gradual degradation of the network in view of redundancy. Conversely, landslide-induced disruptions are spatially small, but they have a disproportionate effect on structurally critical links, thereby causing fragmentation. In addition, the integration of toll roads does not necessarily enhance the robustness of the network when dealing with disruption, indicating an alteration in the structural dependency of the network. These findings support a reframing of road network vulnerability as a structure-conditioned response to disruption, highlighting the significance of structural robustness in infrastructure planning beyond conventional connectivity-based assessments, thereby supporting more effective infrastructure planning and risk mitigation strategies.]]>
