In a new research project between Prof. Dr. Luise Pufahl and Kerstin Andree of the Information Systems Chair, and Dr. Anna Kalenkova, Prof. Dr. Lewis Mitchell, and Wenrui Zhang of the University of Adelaide, we investigate the use of process mining as an explainable data analysis technique in financial business processes. The project is funded by BayIntAn, and will run for the remainder of 2025.
The integration of advanced technologies, particularly artificial intelligence (AI), into financial decision-making processes has driven significant improvements in automation. However, these advancements come with challenges, especially regarding transparency and interpretability. Financial institutions face increasing pressure to ensure their decision-making processes remain efficient, compliant with policies, and transparent to regulators, stakeholders, and customers. As AI models are increasingly deployed in areas such as credit scoring and fraud detection, the demand for clear and comprehensible explanations of their outputs has become critical. Current AI methods often operate as "black boxes," which limits trust and verification capabilities for stakeholders. Process mining, a data analysis field, provides techniques to visualize and formalize workflows based on event log data. These techniques enable the analysis and optimization of business processes, such as loan approval workflows or credit evaluation algorithms. By applying process mining to AI-driven decision-making processes, financial institutions can uncover hidden dependencies, enhance transparency, and redesign their operations.
With this project, we aim to establish a long-term collaboration between the Technical University of Munich (TUM) and the University of Adelaide, Australia. The goal is to research explainable AI with process mining and to apply for DAAD/DFG funding in 2026. The BayIntAn funding program supports us in working together on a first publication and to further strengthen the relationship between the universities by meeting in person.