In recent years, advances in the Internet of Things led to new approaches and applications, for example, in the domains Smart Factories or Smart Cities. However, with the advantages such applications bring, also new challenges arise. One of these challenges is the recognition of situations, e.g., machine failures in Smart Factories. Especially in the domain of industrial manufacturing, several requirements have to be met in order to deliver a reliable and efficient situation recognition. One of these requirements is distribution in order to achieve high efficiency. In this article, we present a layered modeling approach to enable distributed situation recognition. These layers include the modeling, the deployment, and the execution of the situation recognition. Furthermore, we enable tool support to decrease the complexity for domain users.