Complex infrastructure systems, such as transportation and communication networks, consist of many interconnected and interdependent elements that exhibit behavior that is difficult to predict. Researchers have developed System of Systems (SoS) approaches to model and understand the behavior of such complex systems. However, creating SoS models requires combining multiple complex models to describe the system s overall behavior, leading to many open questions about the influence of specific parameters on the results, how uncertainties propagate through the system, and what level of model detail is required. To address these challenges, this paper proposes a network approach to study the properties of computational SoS models for infrastructure systems. The proposed methodology utilizes network analysis to categorize the system and identify critical components, providing insights into the complexity and uncertainty of the system. Furthermore, theoretical concepts from network analysis are used to demonstrate how uncertainties propagate through the system and identify critical model parameters. The approach is applied to a multi-hazard transportation problem for a road network located in Switzerland. The results show that the network representation of SoS models provides a powerful tool for analyzing the behavior of infrastructure systems and can provide insights into how uncertainties propagate and which system components are critical. This approach can help researchers derive further knowledge about complex systems, provide a means for making informed decisions regarding system design and development, and help gain new insights into the behavior of SoS models.