The public defense of Damir Bilic's licentiate thesis in Computer Science and Engineering
The public defense of Damir Bilic's licentiate thesis in Computer Science and Engineering will take place at Mälardalen University, Västerås Campus, room Zeta and Online/Zoom) at 13:00 on November 3, 2020.
Title: “Managing variability in SysML models of automotive systems”.
Serial number: 295.
The faculty examiner is Associate Professor Thorsten Berger, Chalmers, and the examining committee consists of Professor Ina Schaefer, TU Braunschweig, Adjunct Professor Magnus Larsson, ABB Robotics and Associate Professor Thorsten Berger, Chalmers.
Reserve; Professor Andreas Ryve, Mälardalen University.
Fairly recent, yet rapid, advancements in all domains of technology have brought a plethora of challenges for anyone who is developing software-intensive systems. Consider the automotive domain for example. Cars are complex systems, a single car can have highly complex mechanical,
electrical and electronic systems all at once. Moreover, these systems can be highly configurable. For example, a customer might choose a diesel engine, a petrol engine, an electric motor or even a hybrid mix of them, all for the same car model. Not to mention the additional configuration options for comfort, styling or electronic systems such as autopilots.
To manage the complexity and variability of such software-intensive systems, organizations are often looking into model-based techniques for development. One such technique is model-based systems engineering, where a set of models is developed for the planning, design, analysis and documentation of software-intensive systems. Variability is a factor that is often overlooked in model-based systems engineering. Product line engineering is a technique which is used for the development of highly such configurable systems, however, not necessarily using models.
In this thesis, the research focus was on the management of variability in model-based systems engineering. We have been exploring the combination of model-based systems engineering and product line engineering. Initially, the work started with the evaluation of currently existing tool support to manage variability of engine systems developed at Volvo Construction Equipment.
Secondly, we have evaluated how model-based product line engineering affects different phases of the development process at Volvo CE, i.e. we tried to assess the consequences during requirements engineering, during the design, testing, etc. During these studies, several shortcomings were identified in terms of both tool support and methodology support. Often, variability related issues in the product were not detected until late phases in the development process. Lastly, to address some of the identified challenges, in the thesis, we have developed an approach which can detect variability-related inconsistencies in system models during systems engineering.