Scholarlink

Business processes are fundamental to organizational operations, yet their improvement remains challenging due to the time-consuming nature of manual process analysis. Our paper harnesses Large Language Models to automate value-added analysis, which is mostly done manually, time-consuming, and subjective. Our method is more principled and operates in two phases: decomposing high-level activities into detailed steps to enable granular analysis; and performing a value-added analysis to classify each step according to Lean principles. Our approach was developed using 50 business process models, for which we collect and publish manual ground-truth labels. Our evaluation, comparing zero-shot baselines with structured prompts reveals (a) a consistent benefit of structured prompting and (b) promising performance for both tasks.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Automated Business Process Analysis: An LLM-Based Approach to Value Assessment

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • William De Michele,
  • Abel Armas Cervantes,
  • Lea Frermann

摘要

Business processes are fundamental to organizational operations, yet their improvement remains challenging due to the time-consuming nature of manual process analysis. Our paper harnesses Large Language Models to automate value-added analysis, which is mostly done manually, time-consuming, and subjective. Our method is more principled and operates in two phases: decomposing high-level activities into detailed steps to enable granular analysis; and performing a value-added analysis to classify each step according to Lean principles. Our approach was developed using 50 business process models, for which we collect and publish manual ground-truth labels. Our evaluation, comparing zero-shot baselines with structured prompts reveals (a) a consistent benefit of structured prompting and (b) promising performance for both tasks.