Model-Based Risk Assessment in Multi-disciplinary Systems Engineering

In industrial production systems engineering projects, the work of software managers depends on engineering artifacts coming from multiple disciplines. In particular, it is important to software managers to assess the project risk from the status and evolution of various heterogenous distributed engineering artifacts. Thus, software risk management is most often an error prone and cumbersome task in such projects. To tackle this challenge, we introduce a model-based foundation for risk assessment in multi-disciplinary systems engineering projects. In particular, we build on the recent modeling support for the Automation ML (AML) standard which enables representing data coming from different engineering disciplines as models and employ a linking language to reason on a set of distributed engineering artifacts and their relationships. Based on this pillars, we propose in this paper a dedicated metric suite and measurement support for AML as an important ingredient for efficient risk assessment of heterogenous and distributed engineering data. We evaluate the feasibility of the proposed approach by providing tool support on top of the Eclipse Modeling Framework (EMF) and demonstrate its application with a showcase based on a real-world case study.