The BIoTope project, funded by the German Federal Ministry of Education and Research, ran from 1 June 2019 to 31 December 2022 and was carried out by StoneOne AG (Anaqor) together with the Technical University Clausthal, the University of Mannheim and Wolfsburg AG. The main objective was to create a foundational technology and engineering methodology that allows emergent services to be generated automatically on self‑adaptive system platforms. Unlike conventional approaches, the composition rules are not fixed centrally; instead they can be configured dynamically and on demand, enabling the integration of open systems whose interfaces are not governed by a single semantic model, such as many IoT devices that lack industry standards.

At the core of the solution is an Execution Engine that provides a set of basic operations which can be assembled into executable flows. These flows are interpreted and executed by the engine. They can be produced automatically from a Composition Result or created manually through a graphical composer. The Composition Result contains a sequence of action calls together with an environment that defines the required data types. For every action a matching Service Flow is stored in the Service Registry. By chaining these Service Flows in the order prescribed by the actions, a Meta‑Flow is automatically generated. The Meta‑Flow is executed recursively: each sub‑flow is treated as an independent service, and if a sub‑flow contains further sub‑flows, they are executed in turn. This recursive mechanism allows a dynamically generated flow to be reused as a building block in higher‑level flows, thereby supporting the automatic construction of complex, directly executable service chains.

The platform architecture comprises four main components: the User Requirements Handler (URH), the Self‑adaptive Composition Mechanism (SCM), the Execution Engine (EE) and the Service Registry (SR). The URH captures user requirements either explicitly through user requests or implicitly via sensor monitoring. The SCM automatically composes a software service that satisfies the identified requirements. The EE then executes this composed service, selecting the appropriate service instances, while the SR maintains a knowledge base of all available services on the platform. This modular design ensures that new services can be integrated seamlessly, and that the system can adapt to changing user needs without manual reconfiguration.

Demonstration scenarios were developed to validate the approach. In a mobility scenario, the platform dynamically assembled services to provide real‑time route planning and traffic monitoring, while in a smart‑office scenario it coordinated lighting, HVAC and security services based on occupancy and environmental data. These demonstrations showcased the platform’s ability to integrate heterogeneous services, to compose them on the fly, and to execute the resulting flows reliably. Although the report does not provide explicit quantitative performance figures, the successful execution of these scenarios confirms that the system can handle complex, multi‑service chains in real‑time environments.

StoneOne AG was responsible for the economic exploitation of the project results, particularly the development of the Execution Engine. The Technical University Clausthal contributed expertise in software and systems engineering, the University of Mannheim provided research on enterprise systems, and Wolfsburg AG supplied industrial experience in automotive and manufacturing contexts. Together, the consortium delivered a robust, extensible platform that demonstrates how dynamic, demand‑driven service composition can be achieved on self‑adaptive system platforms, paving the way for future research and commercial applications in the rapidly evolving IoT landscape.