Everyone is talking about Industry 4.0 – but what does this actually mean?
According to the German Industrie 4.0 Working Group, it revolves around “networks of manufacturing resources (manufacturing machinery, robots, conveyor and warehousing systems and production facilities) that are autonomous, capable of controlling themselves in response to different situations, self-configuring, knowledge-based, sensor-equipped and spatially dispersed and that also incorporate the relevant planning and management systems” (Recommendations for implementing the strategic initiative INDUSTRIE 4.0).
If we look at this vision from the perspective of systems and software engineering, we see numerous challenges, particularly if, despite the rapidly increasing system complexity, the crucial system features Safety and Security as well as the User Experience of a system must be guaranteed.
In order to meet these challenges, Fraunhofer IESE is working on scalable systems and software engineering methods. These can be applied efficiently in practice despite the complexity of real-life systems. One special focus is on guaranteeing quality features both by construction and with regard to quality assurance. In order to be able to give this guarantee, IESE is working on various issues along the development lifecycle.
The first question that arises concerns the issue of a suitable process for allowing efficient elicitation and assessment of the requirements on such complex, highly dynamic, and interconnected systems. Based on the requirements, the architecture methods of IESE can then be used to derive suitable system architectures. These allow regarding the entire value chain from the sensors to the Cloud service in an integrated manner. In addition, they contribute to assuring the safety, security, and user experience of these systems by construction.
Due to the use of model-based processes, analysis and simulation of the architectures, in particular, is very important already in early development phases.
Especially from the perspective of quality features – such as safety – this results in numerous challenges for the vision of “Industry 4.0“. On the one hand, adjectives such as “autonomous“ or “self-configuring“ call for a large degree of (artificial) intelligence and adaptivity of the individual systems. The flexible interconnection requirement, on the other hand, results in the situation that at runtime, systems of systems are created dynamically whose structure and overall behavior cannot be predicted – or can only be predicted with difficulty – at the time the single systems are developed. All these are factors that lead to uncertainties, i.e., features are hard to predict and thus lead to great uncertainties with regard to predicting the system behavior to be expected. This contradicts the execution of safety cases, which is centrally based on the assumption of deterministic, predictable system behavior. Fraunhofer IESE meets these challenges with modular safety case processes as well as processes that allow safety cases and monitoring to be performed at runtime in order to account for the dynamics of Industry 4.0 systems.
At the same time, the systems become more open as a result of collaboration with other systems, and thus more vulnerable to security attacks. On the one hand, this has an impact on safety; on the other hand, the elicited data must be safeguarded. Due to the increased interconnection and central usage of the data, more powerful data protection mechanisms are required. The researchers at Fraunhofer IESE are therefore developing data usage control concepts that use so-called security policies to determine who is allowed to see or change which data to which extent and in which place. The use of corresponding mechanisms allows guaranteeing adherence to these policies at runtime. This makes it possible to use the data that are essential for Industry 4.0 intensively, without losing usage control over them.