The paper investigates the vulnerability of complex networks to attacks on vertices and edges, using both real-world networks and various network models. The performance of the networks is measured by the average inverse geodesic length and the size of the largest connected subgraph. Four different attack strategies are employed: removals based on initial degrees and betweenness centrality, as well as recalculated degrees and betweenness centrality during the removal process. The study finds that recalculated degrees and betweenness centrality often lead to more harmful attacks than those based on initial network data, indicating that network structure changes as important vertices or edges are removed. The correlation between degree and betweenness centrality in complex networks is also examined, showing that higher degrees generally correlate with higher betweenness centrality, though this correlation weakens in more clustered networks. The vulnerability of various network models, including Erdös-Rényi (ER), Watts-Strogatz (WS), Barabási-Albert (BA), and clustered scale-free (CSF) networks, is analyzed under vertex and edge attacks. The results highlight differences in vulnerability behavior across different network types, with scale-free networks being more sensitive to vertex removal and WS networks showing unique vulnerability patterns due to their small-world properties.The paper investigates the vulnerability of complex networks to attacks on vertices and edges, using both real-world networks and various network models. The performance of the networks is measured by the average inverse geodesic length and the size of the largest connected subgraph. Four different attack strategies are employed: removals based on initial degrees and betweenness centrality, as well as recalculated degrees and betweenness centrality during the removal process. The study finds that recalculated degrees and betweenness centrality often lead to more harmful attacks than those based on initial network data, indicating that network structure changes as important vertices or edges are removed. The correlation between degree and betweenness centrality in complex networks is also examined, showing that higher degrees generally correlate with higher betweenness centrality, though this correlation weakens in more clustered networks. The vulnerability of various network models, including Erdös-Rényi (ER), Watts-Strogatz (WS), Barabási-Albert (BA), and clustered scale-free (CSF) networks, is analyzed under vertex and edge attacks. The results highlight differences in vulnerability behavior across different network types, with scale-free networks being more sensitive to vertex removal and WS networks showing unique vulnerability patterns due to their small-world properties.