Research

My research focuses on the engineering of complex, long-lasting, software-intensive systems within various application domains in an integrated approach with different disciplines. We develop domain-specific languages (DSLs) and methods for domains such as Ambient Assisted Living, Controlling and Finances, Energy, Health, Manufacturing, Internet of Things (IoT), or Smart Homes. The models created with such languages can be used to develop information systems, assistive systems, and digital twins in an agile model-based way using generators for code synthesis. This website presents selected research topics and related publications. The full list of publications is available on the publications page.

Assistive Systems

Assistive systems provide situational support for human behaviour based on information from previously stored and real-time monitored structural context and behaviour data at the time a person needs or asks for it [HMR+19]. They provide critical functionality, or ensure safety in potentially dangerous work environments. Research is needed, as most of them are restricted to certain use cases which allows for less reusability. We investigate needed context information [MS17], modeling languages used during runtime and to support the engineering [MM13], how to use human behavior goals for MDSE approaches [MRV20] show one possible implementation for using goal models in generated assistive systems [MRZ21], and how to automate the engineering of assistive systems. Investigated domains are, e.g., financial management [MRZ21], smart homes, or production [Mic22].

Selected Publications:

Digital Twins & Digital Shadows

In our understanding, a digital twin of an original system is a software that consists of a set of models of the system, a set of digital shadows [BBD+21], and provides a set of services to use the data and models purposefully with respect to the original system (original definition:[DMR+20], Version 2.0 in [HHMR23], Version 2.2 here).

For DT engineering, software and systems models can be used to support the design process, construction, operation, and end-of-life of related physical objects. Our main use cases are from the production domain, as I am leading the workstream “Foundations of Digital Shadows” within the Cluster of Excellence “Internet of Production”. We are working on the conceptualization of digital shadows, DT architectures, how to perform process prediction with DTs and integrate process mining techniques in MDSE, low-code development platforms for DTs, and the generation of DT cockpits and process-aware DT cockpits from event logs. We have investigated integration challenges for DT systems-of-systems and SE challenges in cyber-physical production systems engineering with industry. We work on privacy-preserving system design of IoT systems, and model-driven IoT app stores.

Selected Publications

Model-Based Software Engineering of Information Systems

Using a small set of models, it is possible to generate large parts of information systems. With MontiGem, the generator framework for information systems [GMN+20], we have created a tool to support such a process in an agile, model-driven way.

Selected Publications

Modeling Law

Domain-specific languages provide means to formulate law and contracts in a computer-understandable way. This allows to verify plans according to contracts [DHH+20], and to generate related IT systems or specific components from law [RMK+21].

Selected Publications

Privacy

Due to the General Data Protection Regulation (GDPR) organizations are obliged to consider privacy throughout the complete development process. In our work, we have created privacy-preserving information systems and demonstrated it on an IoT manufacturing use case [MKM+19], we have created an approach for differential privacy for event logs [MKB+19], and we have shown how MDSE can be used to create privacy-preserving IoT systems [MNRV19].

Selected Publications