Maged Khalil successfully completed his doctorate on the topic of "A design pattern-based approach to systematising the reuse of safety-critical system solutions in automotive applications" under the supervision of Professor Roman Dumitrescu. Congratulations on this achievement!
Summary of the thesis:
The reuse of operationally proven solutions - such as safety mechanisms or architecture designs - is widespread in the development of safety-critical systems in order to increase confidence in the design, reduce development effort and promote compliance with requirements. Nevertheless, this reuse is still largely ad hoc and opportunistic due to a lack of process maturity, structured guidance and appropriate tool support. In addition, there is a lack of explicit criteria for assessing the completeness, quality and maturity of reuse, which makes it difficult to determine the actual impact and effectiveness of such activities.
This doctoral dissertation develops a model-based approach to systematically support the reuse of safety-critical, software-intensive system solutions in practice. The focus is on the structured collection, recording and modelling of all information required for correct reuse - integrated into the tools and artefacts used in industrial practice. This results in an industry-compatible framework that enables consistent and context-aware reuse in everyday development. Based on the user-centred principles of design patterns, a modular, model-based pattern library is defined that combines development artefacts and structured reasoning elements and integrates them with guided instructions in a holistic approach. Using the example of safety mechanisms in the automotive sector, a reusable pattern template is developed and instantiated in two environments: a domain-independent research platform and an automotive modelling framework. The methodology is then generalised and applied to a broad-based reuse problem in a long-term industrial case study at a leading global Tier 1 automotive supplier over several years. A systematic expert survey and tool-supported maturity assessments validate the effectiveness and practicability of the approach. The model-based capture of patterns within their practical context offers several key advantages: it improves the characterisation of the problem space, enables a more precise and practical description of the reuse steps and strengthens the coordination of relevant interdisziplinär artefacts - thus promoting the co-evolution of system design artefacts and artefacts from other disciplines (e.g. safety proofs). In addition, the approach provides a mechanism for the systematic evaluation of the reusability of tool chains as well as for the targeted identification and closure of methodological and tool-specific gaps.
The contributions of this work include a novel operationalisation of design patterns, a cross-domain reuse framework and validated guidelines for organisations that want to systematically establish reuse on an industrial scale, and are thus relevant for both science and practice.
