Article Details

Abstract

Aiming at the problems of single diagnosis dimension, insufficient intelligence of rule engine and poor operability of improvement suggestions in the existing evaluation system, this paper studies and proposes an innovative architecture of intelligent evaluation system based on large language model. Firstly, a four-layer technical system covering the basic layer, data layer, capability layer and application layer is constructed. Secondly, the LoRa method is used to fine-tune the ChatGLM3-6B model, and a dedicated training dataset is constructed based on the offline evaluation report. Finally, an intelligent evaluation platform that can automatically generate diagnostic reports and provide executable recommendations is successfully built. The empirical research shows that compared with the traditional rule system, the system has a significant improvement in the coverage of evaluation indicators, and the implementation score of improvement suggestions has also been improved. It effectively solves the pain points such as unclear path and difficulty in landing measures faced by enterprises in the process of digital transformation, and provides new technical support for the intelligent upgrading of manufacturing industry.

Keywords

digital transformation; large model fine-tuning; LoRa

References

1. [1] Ma Jie, Sun Wangchun, Wang Pinghui, etc Knowledge Graph Q&A Based on Relevance Tips [J/OL]. Journal of Software, 2025:1-16 [2020-02-15] https://doi.org/10.13328/j.cnki.jos.007247.
2. [2] Sichuan Provincial Department of Economy and Information Technology Notice on Issuing the Overall Framework of Evaluation Index System for Intelligent Transformation and Digital Transformation of Manufacturing Industry in Sichuan Province (Trial) [EB/OL]. [2024-12-19] https://jxt.sc.gov.cn//scjxt/jxtz/2023/11/8/3717a48f3eb44b808b7a04906294d236.shtml.
3. [3] China Electronics Standardization Institute Intelligent Manufacturing Evaluation Public Service Platform [EB/OL]. [2022-12-19] https://www.c3mep.cn/assessment/list?subPlatformId=1.
4. [4] Ding Peigeng, Liu Zhiwei, Wang Jinqiao Overview and Prospects of Multimodal Artificial Intelligence Large Models [J/OL]. Radio Engineering. 2025:1-20 [2022-12-19] https://link.cnki.net/urlid/13.1097.tn.20241121.1402.002.
5. [5] Ren Lei, Wang Haiteng, Dong Jiabao, etc Industrial Big Model: System Architecture, Key Technologies, and Typical Applications [J]. Chinese Science: Information Science, 2024, 54 (11): 2606-2622
6. [6] Zhang Yunqiu, Yin Ce Research on Entity Automatic Recognition of Chinese Electronic Medical Records Based on Large Models [J/OL]. Data Analysis and Knowledge Discovery, 2025:1-18 [2024-12-19] http://kns.cnki.net/kcms/detail/10.1478.g2.20241118.1735.004.html.
7. [7] Zhang Kun, Zhang Yongwei, Wu Yongcheng, etc Automatic Construction Method of Equipment Fault Knowledge Graph Based on Large Model [J]. Computer and Modernization, 2024 (11): 46-53
8. [8] Liu Zeyuan, Wang Pengjiang, Song Xiaobin, etc A Review of the Illusion Problem in Large Language Models [J]. Journal of Software, 2025, 36 (3): 1152-1185
9. [9] ZENG A H，LIU X，DU Z X，et al. GLM-130B: An Open Bilingual Pre-Trained Model [EB/OL].[2024-12-19].https://arxiv.org/pdf/2210.02414.
10. [10] Wang Shuai, He Wenchun, Wang Fudi, etc A Question Answering System Integrating Knowledge Graph and Vector Retrieval with Large Language Model [J]. Science, Technology and Engineering, 2024, 24 (32): 13902-13910
11. [11] HU E J，SHEN Y L，PHILLIP W，et al. LoRa: Low-Rank Adaptation of Large Language Models [J/OL].arXiv:2106.09685 [cs.CL].[2024-12-19].https://arxiv.org/abs/2106.09685.
12. [12] Ren Haiyu, Liu Jianping, Wang Jian, etc A review of research on intelligent question answering systems based on big language models [J]. Computer Engineering and Applications, 2025, 61 (7): 1-24

Funding

Innovation and Development Project of the Western Branch of China Academy of Information and Communications Technology (2024ZY2872)

About the authors

Zhao Bin (1990-), male, Han nationality, from Nanchong, Sichuan Province, information system project manager, master, research direction: industrial Internet, industrial artificial intelligence; Xiong Shuhan (1989-), female, Han nationality, from Mianzhu, Sichuan Province, engineer, master, research direction: intelligent manufacturing and industrial Internet.