=====Workflow Intelligence, Process Mining, and Explainable AI Systems====

This research area focuses on the analysis, modeling, and verification of behavioral processes in software systems, AI-driven environments, intelligent infrastructures, and autonomous agents. Modern digital systems continuously generate event logs and execution traces viewed as the “behavioral fingerprints” of complex systems, capturing how processes, decisions, and interactions evolve over time.

The research explores how event logs, execution traces, and AI-generated workflows can be transformed into analyzable behavioral models using workflow intelligence and process mining techniques. Particular attention is given to workflow discovery, behavioral analysis, process verification, anomaly detection, and explainability of multi-step decision-making processes generated by AI systems and autonomous agents operating in dynamic environments.

An important research direction concerns semantic and behavioral drift analysis, including detection of evolving patterns, inconsistencies, unstable decision paths, and hidden process anomalies. This is especially relevant for modern LLM-driven agents and autonomous AI systems whose behavior may gradually diverge from expected workflows, organizational rules, or safety constraints. The research investigates methods for monitoring and supervising evolving AI behaviors using workflow-aware analysis and process diagnostics.

The area also includes explainable AI mechanisms based on process analysis, attribution methods, and workflow-oriented reasoning. Special attention is devoted to cooperative game-theoretic approaches, including Shapley-based attribution analysis, for assessing contribution, criticality, and behavioral risk of workflow components and AI decision paths.

Expected outcomes include workflow-aware AI analysis methods, explainable behavioral modeling, intelligent process diagnostics, attribution-based reasoning frameworks, and environments supporting trustworthy operation of AI-driven and software-intensive systems.

=== See also ===

  * **Radoslaw Klimek**, Julia Witek. [[https://dl.acm.org/doi/10.1145/3695750.3695822|Automatic Generation of Logical Specifications for Behavioural Models]]. 39th IEEE/ACM International Conference on Automated Software Engineering Workshops (ASEW-RENE 2024), Sacramento, California, USA, 2024.
  * **Radoslaw Klimek**, Jakub Blazowski. [[https://arxiv.org/abs/2512.09562|Explainable Verification of Hierarchical Workflows Mined from Event Logs with Shapley Values]]. ArXiv, 2025.
  * van der Aalst W.M.P. [[https://link.springer.com/book/10.1007/978-3-662-49851-4|Process Mining: Data Science in Action]]. Springer, 2016.
  * Lundberg S.M., Lee S.-I. [[https://dl.acm.org/doi/10.5555/3295222.3295230|A Unified Approach to Interpreting Model Predictions]]. Advances in Neural Information Processing Systems (NeurIPS 2017), 2017, pages. 4765–4774.
  * Luo J. et al. [[https://arxiv.org/abs/2503.21460|Large Language Model Agent: A Survey on Methodology, Applications and Challenges]]. ArXiv, 2025.