Bachelor and Master Theses

Indutrial systems are complex, with a long life span, and in general business critical. As a consequence, their development needs to adhere well-defined processes to enable effective design, implementation, maintenance, etc. In this context, the International Council on Systems Engineering (INCOSE) sees the future of systems engineering as "predominantly model-based" in its vision for 2035. The reasons behind this vision are improved communication among the multitude of stakeholders involved in system development, earlier verification and validation opportunities, and an increased possibility to automate tasks, to mention a few. SysML is a de facto standard system modelling language widely used in industry. It originates as a specialization of the UML and it has grown more and more as a separate language for systems engineering. Given this growth and the needs for tailoring it even more as a proper systems engineering language, in the latest 3-4 years there has been a large effort to create a brand new version of the language called SysML v2 (as opposed to the previous version history, namely SysML v1.x). Since SysML v2 will be the language to use in the (near) future, it is expected a growing need of migration support, that is tools, as much automated as possible, translating models defined with the older versions of the language to models valid and loadable with the new version of the language. 

SysML v2 is both conceptually and practically a radically different language that the old versions; the language specification task force has been carefully taking into account migration issues, however the available documentation only provides generic guidelines and correspondences between the language versions. Moreover, industrial models might include several hundreds of elements, often created in domain-specific languages tailored to a specific company and partially including SysML concepts. Given this context, this thesis work explored automation opportunities to migrate models conforming to the older versions of SysML towards the new SysML language specification.

Thesis main goals:

- classify migration problems and the potential automatization support based on sub-portions of the language taken into account (e.g. fully automatic, semi-automatic, not automatic);

- propose a prototypical implementation of migration algorithms;

- validate the migration solution on a selected dataset.

REQUIRED:

- experience with software modelling in UML or similar;

- experience with programming languages (Java and/or Python);

- experience with Jupyter notebooks and/or Eclipse;

OPTIONAL:

- completed a course on software modelling, model-based development, or similar.

This thesis is also suitable for 2 students.

As mentioned in the description this topic is highly relevant for companies we usually collaborate with. In this respect, we might be able to involve companies e.g. by getting their models as validation scenarios for the proposed migration solutions.