Co-Construction of Production-Line AAS Geometric Models and Knowledge Graphs Driven by a Vision-Language Model
DOI:
https://doi.org/10.54097/ht37b711Keywords:
Asset Administration Shell (AAS), Vision-language Model, Knowledge Graph, Digital Twin, Collaborative Modeling, Industry 4.0Abstract
Digital twin systems for a production line usually run on two models at once: an Asset Administration Shell (AAS) that holds asset information under IEC 63278, and a knowledge graph (KG) that records device topology and process semantics. On real projects, though, the two are built apart from each other. Engineers key the AAS geometric model in by hand from drawings and field measurements, while process engineers author the KG separately. The same data therefore gets entered twice, reuse is poor, and sooner or later the two models disagree. This paper presents a co-construction method that removes the duplication at its source. A fine-tuned vision-language model (VLM) reads the engineering sources a line already has, namely 2D/3D CAD drawings, on-site laser point clouds, and equipment technical manuals, and fills a standardized AAS; the KG is then derived from that AAS by fixed rules, so both models share one source. The method is arranged in four layers: multi-source data preprocessing with coarse-to-fine ICP spatial registration; LoRA-based domain adaptation of an open-source VLM supported by industrial prompt templates; automatic population of the AAS geometric, property, and connectivity submodels with conformance-checked .aasx generation on Eclipse BaSyx, plus QEM-based mesh simplification; and same-source KG derivation through fixed field-to-triple mapping rules, backed by a single-source constraint, automatic reconciliation, and incremental refresh with version traceability. We demonstrate the complete pipeline on a photovoltaic cell production line, and a parameterized Monte Carlo simulation of data consistency shows that, for manual single-entry error rates between 1% and 10%, conventional dual-entry modeling produces 4–39 times more data inconsistencies than the proposed single-source derivation, with the gap widening as the error rate grows. The method offers a reusable and extensible path toward production-line digital twin modeling; empirical validation on a physical line is planned as future work.
Downloads
References
[1] Tao, F., Zhang, H., Liu, A., & Nee, A. Y. C. (2019). Digital twin in industry: State-of-the-art. IEEE Transactions on Industrial Informatics, 15(4), 2405–2415. https://doi.org/10.1109/TII.2018.2873186. DOI: https://doi.org/10.1109/TII.2018.2873186
[2] Tao, F., Liu, W. R., Zhang, M., Hu, T. L., Qi, Q. L., Zhang, H., Sui, F. Y., Wang, T., Xu, H., Huang, Z. G., Ma, X., Zhang, L. C., Cheng, J. F., Yao, N. K., Yi, W. M., Zhu, K. Z., Zhang, X. S., Meng, F. J., Jin, X. H., Li, Y., Lv, Q., & Luo, Y. M. (2019). Five-dimension digital twin model and its ten applications. Computer Integrated Manufacturing Systems, 25(1), 1–18. [In Chinese]. https://doi.org/10.13196/j.cims.2019.01.001.
[3] Tao, F., Cheng, Y., Cheng, J. F., Zhang, M., Qi, Q. L., & Zhang, H. (2017). Theory and technology of cyber-physical fusion in digital twin shop floors. Computer Integrated Manufacturing Systems, 23(8), 1603–1611. [In Chinese].
[4] International Electrotechnical Commission. (2023). IEC 63278-1:2023. Asset administration shell for industrial applications—Part 1: Asset administration shell structure. IEC.
[5] Tao, J., Dai, Y. C., & Wei, R. (2017). Research on production life cycle based on digital thread and digital twin. Aeronautical Manufacturing Technology, 60(21), 26–31. [In Chinese].
[6] Hogan, A., Blomqvist, E., Cochez, M., d’Amato, C., de Melo, G., Rigotti, C., Saleem, M., & Vrandecic, D. (2021). Knowledge graphs. ACM Computing Surveys, 54(4), Article 71, 1–37. https://doi.org/10.1145/3447772. DOI: https://doi.org/10.1145/3447772
[7] Liu, X. B., Wang, S. T., & Bai, C. Y. (2022). Visualization analysis of digital twin development based on a scientometric knowledge map. Computer Integrated Manufacturing Systems, 28(6), 1673–1684. [In Chinese].
[8] Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., Krueger, G., & Sutskever, I. (2021). Learning transferable visual models from natural language supervision. In Proceedings of the 38th International Conference on Machine Learning (pp. 8748–8763). PMLR. https://doi.org/10.48550/arXiv.2103.00020.
[9] Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., Xiao, T., Whitehead, S., Berg, A. C., Lo, W. Y., Dollár, P., & Girshick, R. (2023). Segment anything. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 4015–4026). IEEE. https://doi.org/10.1109/ICCV51070.2023.00371. DOI: https://doi.org/10.1109/ICCV51070.2023.00371
[10] Wang, S. B. (2025). Research on knowledge graph evolution based on multimodal large models. China High-Tech, (2), 25–27. [In Chinese].
[11] Liu, Q. W., & Yang, G. (2025). Application research and exploration of digital twins in the machinery field under Industry 4.0. Intelligent Manufacturing, 32(3), 95–100. [In Chinese].
[12] Guo, L., Zhang, H. Q., Cheng, W. S., & Liu, Y. (2023). Research status and prospects of data-driven digital twin technology. Machinery, 50(7), 1–10. [In Chinese].
[13] Rauh, L., Reichardt, M., & Schotten, H. D. (2022). AI asset management: A case study with the asset administration shell (AAS). In 2022 IEEE 27th International Conference on Emerging Technologies and Factory Automation. IEEE. https://doi.org/10.1109/ETFA52439.2022.9921705. DOI: https://doi.org/10.1109/ETFA52439.2022.9921705
[14] Cavalieri, S., & Salafia, M. G. (2020). Insights into mapping solutions based on OPC UA information model applied to the Industry 4.0 asset administration shell. Computers, 9(2), 28. https://doi.org/10.3390/computers9020028. DOI: https://doi.org/10.3390/computers9020028
[15] Zhou, H. T. (2024). Design and implementation of a flexible distributed motion control component based on AAS [Master’s thesis]. Huazhong University of Science and Technology. [In Chinese].
[16] Tao, F., Liu, W. R., Liu, J. H., Liu, X. J., Liu, Q., Qu, T., Hu, T. L., Zhang, Z. N., Xiang, F., Xu, W., & Zhang, H. (2018). Digital twin and its application exploration. Computer Integrated Manufacturing Systems, 24(1), 1–18. [In Chinese].
[17] Banerjee, A., Dalal, R., Mittal, S., & Joshi, S. (2017). Generating digital twin models using knowledge graphs for industrial production lines. In Workshop on Industrial Knowledge Graphs, co-located with the 9th ACM Web Science Conference. DOI: https://doi.org/10.1145/3091478.3162383
[18] Buchgeher, G., Gabauer, D., Martinez-Gil, J., & Bergner, K. (2021). Knowledge graphs in manufacturing and production: A systematic literature review. IEEE Access, 9, 55537–55554. https://doi.org/10.1109/ACCESS.2021.3070395. DOI: https://doi.org/10.1109/ACCESS.2021.3070395
[19] Cao, Y. Y., Xu, Z., Tian, Z. G., & Zhang, L. (2024). Research on a digital twin knowledge graph for manufacturing workshops. Mechanical Science and Technology for Aerospace Engineering, 43(12), 2089–2095. [In Chinese]. https://doi.org/10.13433/j.cnki.1003-8728.20230128.
[20] Elshafei, B., Martínez-Arellano, G., Chaplin, J. C., & Ratchev, S. (2024). Semantic knowledge representation in asset administration shells: Empowering manufacturing utilization. In Flexible Automation and Intelligent Manufacturing (FAIM 2024) (pp. 282–296). Springer. https://doi.org/10.1007/978-3-031-74482-2_32. DOI: https://doi.org/10.1007/978-3-031-74482-2_32
[21] Wang, X., Zheng, J., Hu, Y., Li, Y., & Zhang, H. (2025). From 2D CAD drawings to 3D parametric models: A vision-language approach. Proceedings of the AAAI Conference on Artificial Intelligence, 39(8), 7961–7969. https://doi.org/10.1609/aaai.v39i8.32858. DOI: https://doi.org/10.1609/aaai.v39i8.32858
[22] Garland, M., & Heckbert, P. S. (1997). Surface simplification using quadric error metrics. In Proceedings of the 24th Annual Conference on Computer Graphics and Interactive Techniques (pp. 209–216). DOI: https://doi.org/10.1145/258734.258849
[23] Hu, E. J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., & Chen, W. (2022). LoRA: Low-rank adaptation of large language models. In Proceedings of the 10th International Conference on Learning Representations. https://doi.org/10.48550/arXiv.2106.09685.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Frontiers in Computing and Intelligent Systems
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
