Working Group

Model-Based Assistance and Information Services (MBAIS)


MBAIS - Model-Based Assistance and Information Services Informatics as a research discipline must constantly face new challenges such as heterogeneous information sources, growing amounts of structural and behavioural data, and broad range of possible end-users. Thus, the need for (a) assistive and (b) information services and systems grows. We meet these challenges by using models and model-based software engineering as foundation to develop data-centric information systems.

Model-Based Software Engineering. Generating software is an important capability for a tooling infrastructure that successfully assists modeling in the development process (see "Agile Model Based Software Engineering"). As every application domain has its own terms, relations, structure and behaviour, there is a need for developing different Domain Specific Languages (DSLs) for each domain and/or purpose. The automatic combination of models from a variety of DSLs ensures consistency (e.g. for front- and backend).

Data-Centric Information Systems and Services. Systems and services for the representation of big amounts of data such as web-information portals are commonly used and their representation and system architecture highly depends on the domain in focus. Using generative methods accelerates the development process (see "Generative Software Engineering") and decreases development time, as changes of the models are easily passed on to the source code. We are demonstrating the practical application of this approach, e.g., in the MaCoCo project.

Human-Centered Assistance Systems and Services. A step further from informing users is to provide support to them. Supportive or also called assistive systems and services support human processes both, at work and in private life. It is again an emerging research field regarding the digital transformation of numerous application areas. Such assistance services are commonly embedded in a larger system architecture combining devices and resources, receiving data from sensors and communicate with functions and services. Model-based approaches improve the communication capabilities of these components and ensure consistency. Formal verification methods can be applied to a safety-critical subset of these components and their communication. Moreover, model-based approaches can be used to support privacy-by-design and eases human-centered development approaches.

Application Domains. Assistance and Information Services are relevant in a variety of domains. Each of them with its' own specifications, ideas and legal restrictions. Thus, it is important to strongly involve domain experts into the development process of such systems and services. This involvement can be supported by using MBSE: Prototypes and functionalities can be developed fast, changes can be easily handled, end users can provide feedback in an earlier development stage and feature requests can be implemented faster.

The group members already have experiences in the following domains:

  • Ambient Assisted Living
  • Controlling and Finances
  • Energy
  • Health
  • Internet of Production (IoP)
  • Internet of Things (IoT)
  • Industry 4.0
  • Smart Homes



Research Questions

  • What are further evolvement steps in theoretical foundations of model based code generation? How is it possible to combine different languages?
  • What are the domain specific modelling languages and methods required for the creation of web information systems? How are these engineered, integrated, and reused? How does this translate to modeling tools, analysis, transformations, and code generators?
  • What are the domain specific modelling languages and methods required for assistance services? What are the architectural components and how are they interrelated?
  • How is it possible to ensure privacy-by-design of information and assistive systems in combination by using model-based software engineering? What aspects are relevant to be represented in privacy information portals to support privacy preservation for end users?
  • How can Artifitial Intelligence help to achieve Digital Quality Assurance?


Team Members

Currently, the Model-Based Assistance and Information Services working group consists of:


Judith Michael, Dr.
(team leader)

Kai Adam, M.Sc.

Lukas Netz, M.Sc.

Deni Raco, Dipl.-Inform.

Simon Varga, M.Sc.


Teaching

Our research activities and their results influence the courses we offer and support. In the past, research in model-based software engineering, software language engineering, and their application in web-information systems manifested in project classes and seminars:


Master Theses

We always have ideas for challenging Bachelor and Master theses:

Ongoing master theses

  • Domain specific adaptation/extension of (domain) models to fit specific needs
  • Generating of Test-Data for modelled data structures
  • Generation of tests by using behavior descriptions described with sequence-diagrams
  • Development of a DSL for Graphical User Interfaces
  • Recording and processing of errors in web applications
  • A generative, state-based methodology for verifying underspecified distributed systems

Open topics for master theses

Currently, we're looking for students willing to investigate the following research topics.

  • User-Centered and Privacy-Driven System Design
  • Data Modeling for Privacy Prevention
  • Modeling Human Behavior for Assistive Systems
  • Goal Modeling for Behavioral Support
  • Model Transformation between different DSLs
  • Artifitial Intelligence for Digital Quality Assurance

Feel free to contact us if your are interested in one of these topics!


Publications

Our research manifests in project results with large and small partners from industry and research, as well as in our publications. You can find our publications by topic:

Modeling

Modeling Languages

Application Domains

Our Most Important Software Tools