Name of participant: Humam Kourani
Project’s name: Streamlined Business Process Modeling (SBPM)
Project description:
Understanding and modeling business processes is a key step toward improving organizational workflows. However, creating formal models can be challenging—especially when information is spread across unstructured documents or buried in large volumes of event data.
The SBPM project (Streamlined Business Process Modeling) explores ways to support this task by developing a prototype that can generate process models from both textual descriptions and event logs. The idea is to combine insights from natural language with patterns observed in real-world data to offer a more complete and accessible modeling experience.
Recent advances in Large Language Models (LLMs) and Process Mining provide promising tools for this kind of approach. In SBPM, we aim to investigate how these technologies can be used in practice—bringing together language understanding and data analysis in a single workflow. The project will also make use of the Partially Ordered Workflow Language (POWL), which helps ensure that the generated models are structurally correct and suitable for further analysis or execution.
The prototype will include:
- Components for automated model generation from both text and data.
- Support for combining these sources to enrich the modeling outcome.
- A user-friendly interface with editing, visualization, and export features.
- Feedback mechanisms to iteratively refine model quality.
SBPM is intended as a research-driven contribution toward more efficient, hybrid modeling tools that can support both technical and non-technical users. By working closely with academic and industry partners, the project also aims to reflect practical needs and explore potential for future applications.
Software Campus Partner: Fraunhofer ICT Group and Celonis SE
Implementation period: 01.03.2025 – 28.02.2027
