Automated BPMN Model Generation from Textual Process Descriptions: A Multi-Stage LLM-Driven Approach
Ion Matei, Maksym Zhenirovskyy, Praveen Kumar Menaka Sekar, Hon Yung Wong
TLDR
This paper introduces a multi-stage LLM pipeline that automates BPMN model generation from text and constructs a validated ground-truth corpus.
Key contributions
- Presents a scalable, multi-stage LLM pipeline for automated BPMN model generation from text.
- Automates ground-truth construction via translation, execution-oriented validation, and LLM-guided repair.
- Reconstructs executable BPMN 2.0 XML diagrams from LLM-generated process descriptions.
- Introduces a multi-dimensional similarity framework for evaluating BPMN model reconstruction.
Why it matters
This paper addresses the long-standing challenge of automatically generating BPMN models from unstructured text, a complex task due to heterogeneous conventions. It demonstrates that LLMs can effectively create structurally compliant and semantically meaningful BPMN diagrams at scale, offering a significant step towards automating process modeling.
Original Abstract
Automatically reconstructing BPMN models from unstructured natural-language descriptions remains challenging due to heterogeneous modeling conventions, multilingual sources, and the lack of reliable ground truth. We present a scalable, multi-stage LLM-driven pipeline that automates both ground-truth construction and model reconstruction. Multilingual BPMN XML files are translated into English, validated using execution-oriented compliance checks in SpiffWorkflow, and iteratively repaired through targeted LLM-guided corrections to produce a consistent ground-truth corpus. From these validated models, process descriptions are generated and used to reconstruct executable BPMN~2.0 XML diagrams without manual curation. We introduce a multi-dimensional similarity framework combining structural metrics, type-distribution alignment, and embedding-based semantic measures. In an empirical study of 750 public BPMN diagrams, the pipeline generated 387 validated ground-truth models and achieved average reconstruction similarity above 0.75, including approximately 50 near-perfect reconstructions differing only in minor naming variations. The results demonstrate that LLMs can generate structurally compliant and semantically meaningful BPMN diagrams at scale.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.