AbstractComputer networks are vital in our everyday lives. It is important to design network configurations with special consideration for their various aspects, such as security, integrity, scalability, and cost. It is especially important for a network to be built as robustly as possible to protect against failures, attacks, and intrusions. In this article, I review methods to assess and improve the robustness and efficiency of computer networks. These methods use computer network analysis, social network analysis, evolutionary computing, statistical methods, and graph theory. Specifically, the aim has been to achieve enhanced network robustness and efficiency with a primary focus on architecture and topology of networks. Metrics have been developed for measuring the robustness and efficiency elements of networks and to construct an evolutionary algorithm for the enhancement of these elements. These methods have been applied to various networks, including random networks, biased networks, and real‐life networks. These networks have been analyzed and enhanced using the evolutionary algorithm. Using the metrics, it is shown how the robustness and efficiency of the networks improve. In addition, through this evolutionary process, certain network parameters, as well as the network topological configuration converge. WIREs Comp Stat 2011 3 34–46 DOI: 10.1002/wics.135This article is categorized under: Algorithms and Computational Methods > Genetic Algorithms and Evolutionary Computing Data: Types and Structure > Graph and Network Data Algorithms and Computational Methods > Networks and Security