Automatic business process modeling method based on large language models

被引：0

作者：

Liu, Ruixiang ^{[1
]}

Liu, Xianhui ^{[1
]}

Zhao, Weidong ^{[1
]}

Zhu, Chenglin ^{[2
]}

机构：

[1] College of Electronic and Information Engineering, Tongji University, Shanghai

[2] COSMOPlat Industrial Intelligence Research Institute, Qingdao

来源：

Jisuanji Jicheng Zhizao Xitong/Computer Integrated Manufacturing Systems, CIMS | 2025年 / 31卷 / 06期

关键词：

automated modeling; business process modeling; large language models; loop structure recognition; multi-agent;

D O I：

10.13196/j.cims.2024.0399

中图分类号：

学科分类号：

摘要：

To improve the automation of business process model construction in projects or enterprises and effectively utilize unstructured data within processes, a method for automated business process modeling based on Large Language Models (LLMs) was proposed. In this modeling framework, a Large Language Model was employed to construct an input collection module that accepted inputs in natural language text and other unstructured formats. A simple markup language was used to intermediate represent the business processes. Sequential transformation rules were constructed to fine-tune the Large Language Model's learning on the usage of the markup language, facilitating the conversion from natural language inputs to intermediate representation. Additionally, to correct grammatical errors in results and enhance modeling outcomes, a multi-agent correction module was introduced to automatically inspect and rectify results. Finally, modeling experiments were conducted on business processes with different characteristics. validating the automation of the proposed modeling method and its effectiveness in extracting implicit loop structures. © 2025 CIMS. All rights reserved.

引用

页码：2001 / 2014

页数：13

共 33 条

[1]

DE CAMARGO J V, DE MOREIRA BOHNENBERGER N M, STEIN DANI V, Et al., A complementary analysis of the behavior of BPMN tools regarding process modeling problems, Proceedings of Enterprise, Business-Process and Information Systems Modeling, pp. 43-59, (2022)

[2]

FRIEDRICH F, MENDLING J, PUHLMANN F., Processmodel generation from natural language text, Proceedings of Advanced Information Systems Engineering, pp. 482-496, (2011)

[3]

ZHOU G R, ZHU X Q, SONG C R, Et al., Deep interest network for click-through rate prediction, Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery Data Mining, pp. 1059-1068, (2018)

[4]

ZHAO Haiyan, FU Jianping, GUAN Wei, Et al., Ensemble learning-based business process anomaly detection and localization framework [J/OL], Computer Integrated Manufacturing Systems, pp. 1-15, (2024)

[5]

KO J., COMUZZI M., Asystematic review of anomaly detection for business process event logs [J], Business & Information Systems Engineering, 65, 4, pp. 441-162, (2023)

[6]

XU J D., QI M H., Business process modeling extension and schedu ling strategy compatible with big data business[C], Proceedings of 2023 IEEE International Conference on Control. Electronics and Computer Technology, pp. 827-831, (2023)

[7]

AMJAD A., AZAM F., ANWAR M W., Et al., Event-driven process chain for modeling and verification of business requirements A systematic literature review [J], IEEE Access, 6, pp. 9027-9048, (2018)

[8]

HUANG Fenglan, Feng NI, LIU Jiang, Et al., Data flow modeling and verification of complex BPMN collaboration models based on HCPN[J], Computer Integrated Manufacturing Systems, 30, 5, pp. 1754-1769, (2024)

[9]

WANG Yuqi, MO Qi, WANG Jianeng, Et al., Public process-o-riented correctness verification approach for collaborative business processes based on Petri nets [J], Computer Integrated Manufacturing Systems, 30, 8, pp. 2854-2871, (2024)

[10]

FARSHIDI S., KWANTES I B., JANSEN S., Business process modeling language selection for research modelers [J], Software and Systems Modeling, 23, 1, pp. 137-162, (2024)

← 1 2 3 4 →