Analyzing the Impact of Vehicle Node Survivability Using the Graph Invariant Methodology in Smart Cities

The existence of complex networks is prevalent across various domains, including transportation systems, underscoring the necessity of understanding their resilience for maintaining operational stability. This study examines the impact of vehicular node failures, arising from targeted attacks or random incidents, on the stability of centrality metrics across diverse graph topologies, including Star, Ring, Partial Mesh, Trimet, Dual Ring, Tree, Balanced Tree, and Hybrid Tree. A comparative analysis was conducted using centrality measures, such as degree, betweenness, closeness, eigenvector centrality, clustering coefficient, and assortativity, employing a Python-based simulation framework to model the node failures at rates of 10%, 20%, and 30%. The results indicate that the differential impacts of the node failures vary based on the centrality metric and network topology; notably, the degree centrality demonstrates greater resilience, while the betweenness centrality is particularly sensitive to disruptions. Additionally, the network size, average degree, and the nature of the node failures significantly influence these metrics' stability. The findings highlight the critical relationships between the node survivability and centrality stability, offering valuable insights for developing robust Vehicular Ad Hoc Networks (VANETs) and guiding strategic decision-making in transportation planning and infrastructure development.