The push for digitalization in the production industry, itself driven by the need for sustainable practices and efficient resource utilization, necessitates effective data management and utilization strategies. An ambitious research project in service of this goal, the Internet of Production, envisions the creation of a network of interconnected production systems akin to a “World Wide Lab.” The project brings together expertise and experimental setups from various manufacturing technologies, material science, and computer science. A central concept is the Digital Shadow, a virtual representation of physical entities or processes, offering valuable insights for data-driven decisions. However, the integration of diverse domain-specific data formats and disciplinary perspectives, ranging from machine to system-level processes, presents a significant challenge. This paper introduces “Models-in-the-Middle” as an approach that calls for a limited number of intermediary minimal data models between data source and data science processing. Building on the success of Object-Centric Event Data (OCED) in addressing the Model-in-the-Middle between heterogeneous enterprise information systems and specialized process mining analytical tools, we present a proposed data format called Measurement and Event Data (MAED), specifically created for IoT data. We present applicable analysis approaches for data instances of these models and discuss how they, together with other such models, can help not only with data management but also with efficient utilization.