From the above calculation of the complexity and connection probability of ER random networks with different node sizes, we can see that the relationship between the complexity value and the connection probability of ER random networks satisfies the same characteristics and is exponential as the connection probability increases. The increase reaches a maximum value, and this maximum value is obtained when the connection probability is approximately 0.17. Subsequently, as the connection probability increases, the complexity value decreases exponentially to a steady state value. The connection probability that reaches the steady state value is obtained again when the connection probability value is approximately 0.42. This phenomenon has great significance: Theoretically speaking, Erds and Rnyi proposed to discuss the maximum and minimum degree distributions in random graphs, and Boll-boas deduced that all degree distributions follow the Poisson distribution, as well as the complexity of ER random networks. And the connection probability From the simulation results, we get similar changes in the law. The connection probabilities of the maximum point and the steady-state value point all have approximate values, but they do not change as the size of the network increases. This is also consistent with the definition of Q- property proposed by Erds and Rnyi. The laws of these simulations have yet to be further deduced theoretically. On the other hand, it is also of great significance to the practice of disruption of supply chain risk. In the transmission of supply chain interruption risks, the complexity of the network obviously affects the robustness of the supply chain network and the spread of risk. The interruption of supply chain risk is equivalent to the deletion of the network side. Therefore, the study of the interruption of the risk domain value is an effective basis for understanding the risk propagation rules of the supply chain network and the risk control and prevention.