Documenting SME Processes with Conversational AI: From Tacit Knowledge to BPMN

Introduces conversational LLMs to streamline the documentation of business processes for Small and Medium-sized Enterprises (SMEs), transforming tacit knowledge into formal BPMN diagrams to enhance operational transparency and efficiency. This directly benefits SMEs by automating and standardizing critical business process documentation.

The paper addresses a critical bottleneck in the digitalization of Small and Medium Enterprises (SMEs): the high cost and complexity of extracting 'tacit knowledge' into formal Business Process Management (BPM) standards. By leveraging Large Language Models (LLMs) as conversational agents to conduct interviews and generate BPMN 2.0 XML, the authors propose a scalable solution to a traditionally manual, consultant-heavy task. The strength of the approach lies in its democratizing potential, allowing non-experts to document workflows. However, the analysis reveals significant risks regarding the 'hallucination' of process steps and the handling of edge cases. While LLMs excel at standardizing text, their ability to infer complex logical gateways (XOR/AND splits) from ambiguous human speech remains prone to error without rigorous human-in-the-loop validation. Furthermore, the paper could better address data privacy concerns, which are paramount when uploading proprietary operational data to cloud-based AI models.