[Back]


Doctor's Theses (authored and supervised):

P. Kalb:
"Continuous Software Quality Management for Evolutionary Systems";
Supervisor, Reviewer: G. Kappel, R. Breu; Institut für Softwaretechnik und Interaktive Systeme, 2016; oral examination: 2016-11-30.



English abstract:
Current trends such as ubiquitous cloud services and complex cyber-physical systems
show that software has emerged from isolated entities to distributed and inter-linked
systems across organizational boundaries. All artefacts of such a software system are
continuously evolving. This brings also new challenges for software engineering which
deals with the adherence to a certain level of quality in such a system.
Today, an increasing community propagates the use of integrated model repositories
for interchanging, interlinking and analysing software engineering data. Furthermore,
ticket based systems are used to coordinate actions of manifold stakeholders working
on this data. Still, both options do not provide sufficient solutions to all of the current
challenges.
In this thesis we will define a novel model-based methodology to conciliate model
repositories and task coordination systems. This method is based on a Common Meta
Model for all heterogeneous artefacts to establish a proper linking and management
of these. On top of the Common Meta Model we will present a state machine based
workflow language that enables control of system evolution by workflows triggered by
system changes. The language fosters orchestration of automated and manual actions
on the repository in a change-driven way. Besides the methodology we will also prototype
a repository able to store and manage instances of the Common Meta Model. We
will apply the methodology on two case studies and derive application patterns for the
workflow language.

German abstract:
Current trends such as ubiquitous cloud services and complex cyber-physical systems
show that software has emerged from isolated entities to distributed and inter-linked
systems across organizational boundaries. All artefacts of such a software system are
continuously evolving. This brings also new challenges for software engineering which
deals with the adherence to a certain level of quality in such a system.
Today, an increasing community propagates the use of integrated model repositories
for interchanging, interlinking and analysing software engineering data. Furthermore,
ticket based systems are used to coordinate actions of manifold stakeholders working
on this data. Still, both options do not provide sufficient solutions to all of the current
challenges.
In this thesis we will define a novel model-based methodology to conciliate model
repositories and task coordination systems. This method is based on a Common Meta
Model for all heterogeneous artefacts to establish a proper linking and management
of these. On top of the Common Meta Model we will present a state machine based
workflow language that enables control of system evolution by workflows triggered by
system changes. The language fosters orchestration of automated and manual actions
on the repository in a change-driven way. Besides the methodology we will also prototype
a repository able to store and manage instances of the Common Meta Model. We
will apply the methodology on two case studies and derive application patterns for the
workflow language.

Keywords:
cloud services, complex cyper physical systems

Created from the Publication Database of the Vienna University of Technology.