Organizations are increasingly looking to adopt the Internet of Things (IoT) to collect the data required for data-driven decision-making. IoT might yield many benefits for asset management organizations engaged in infrastructure asset management, yet not all organizations are equipped to handle this data. IoT data is collected, stored, and analyzed within data infrastructures and there are many changes over time, resulting in the evolution of the data infrastructure and the need to view data infrastructures as complex adaptive systems (CAS). Such data infrastructures represent information about physical reality, in this case about the underlying physical infrastructure. Physical infrastructures are often described and analyzed in literature as CASs, but their underlying data infrastructures are not yet systematically analyzed, whereas they can also be viewed as CAS. Current asset management data models tend to view the system from a static perspective, posing constraints on the extensibility of the system, and making it difficult to adopt new data sources such as IoT. The objective of the research is therefore to develop an extensible model of asset management data infrastructures which helps organizations implement data infrastructures which are capable of evolution and aids the successful adoption of IoT. Systematic literature review and an IoT case study in the infrastructure management domain are used as research methods. By adopting a CAS lens in the design, the resulting data infrastructure is extendable to deal with evolution of asset management data infrastructures in the face of new technologies and new requirements and to steadily exhibit new forms of emergent behavior. This paper concludes that asset management data infrastructures are inherently multilevel, consisting of subsystems, links, and nodes, all of which are interdependent in several ways.