Industrial companies are currently competing for higher degrees of automation and adaptability in their factories. In the course of the fourth industrial revolution factories have become smarter by the introduction of technologies such as cyber-physical systems, Cloud Computing, 5G, and the Internet of Things. With the introduction of systems into their factories that combine these technologies such as Driverless Transport Systems, industrial companies are confronted with a previously unknown web of complexity that emerges from the interconnectivity between the various application components of these systems. The amount of collected data and the need for data analysis in industrial environments grows steadily and results in a new accentuated role for the IT. Due to this trend, system architects are now often faced with large and heterogeneous environments when introducing a new system into an existing smart factory and have to look attentively at the way they embed new applications into the existing IT-infrastructure. To reduce the cost in general, the goal is to reuse existing computation resources for new applications where possible. Finding optimal placement locations for multiple application components in a pool of resources that ranges from traditional options like on-device computing capabilities to emerging options like the edge cloud, is an optimization problem that is often described as the Application Component Placement (ACP) problem. This thesis addresses these issues by presenting a conceptual approach to solving this problem for an industrial use case by making use of the concepts of the Topology and Orchestration Specification for Cloud Applications (TOSCA) standard and extending the capabilities of the OpenTOSCA ecosystem by providing a prototypical implementation of an ACP-solving algorithm.