The study is devoted to a comprehensive study of the influence of the value of links between components of intelligent systems on the sustainability and security of the critical information infrastructure of a smart city. A set of scientific and methodological approaches and formal mathematical models for the design, optimization and management of critical information infrastructure with a focus on maximizing the synergistic effect of interactions is proposed. A methodology for compositional modeling of the architecture of critical information infrastructure based on the concept of the Internet of Things (IoT) has been developed, providing multi-level abstraction through the aggregation of models taking into account interdependencies. A discrete model of the states of a critical node of the network operations center has been built. The synergistic efficiency of the integration of IoT, machine learning and blockchain technologies has been proven to improve security, proactivity and data integrity. To solve the problems of optimizing the configuration, resource allocation and placement of services of the critical information infrastructure, the following have been developed, parameterized and verified: the formalism of Markov decision processes, the method of structural optimization of neural networks, genetic algorithms and the particle swarm method with given parameters. The proposed approaches ensure scalability, interoperability and energy efficiency of the critical information infrastructure, maximize the value of functional and information links, and significantly increase the sustainability, security and overall efficiency of the critical information infrastructure of a smart city, which is a key factor in its sustainable development.