Evaluation of working documentation using digital models and neural network classification

Introduction. The digitalization of the construction industry and the introduction of information modelling (TIM) technologies require the development of new approaches to the evaluation of working documentation. Traditional methods of manual verification are time-consuming, subjective and do not provide reproducible results. This paper proposes a method for intelligent evaluation of working documentation prepared using information modelling tools, based on formalized logical rules and neural network analysis.

Materials and methods. The methodology implements a two-channel approach: parallel assessment of the digital model and textual and graphic documentation. It is based on a multi-level structure of indicators, a logical Boolean model, as well as a neural network architecture that includes a graph subnet (GNN), a text subnet (BERT) and a convolutional subnet (CNN), combined into a multilayer classifier. As a result, there are four discrete decisions: adopted, adopted with revision, sent for revision, refusal to accept. The possibility of working with an incomplete set of documentation is taken into account. Verification of the methodology was carried out using an expert survey.

Results. A mathematical model has been developed that describes the logic of evaluating documentation by performance criteria, percentage compliance and quantitative metrics. Expert validation showed a high consistency of assessments (W ≈ 0.52), especially in the logic of the structure, division into critical groups and two-channel. The most problematic aspects are the architecture of the neural network and feedback.

Conclusions. The methodology proved its applicability for the tasks of internal audit, automation of control over the acceptance of documentation, preparation for examination and assessment of the degree of readiness of working products. The development of
the model is possible through the clarification of the architecture and the expansion of the set of indicators.

1. Kryukov K.M. Practical approach to integration of building information modeling and lean construction. Journal of Construction and Architecture. 2025; 27(1):180-193. DOI: 10.31675/1607-1859-2025-27-1-180-193. EDN SRJTJS. (rus.).

2. Virtsev M.Yu., Salakhov R.L. Advantages and scope of application BIM technologies in Russia. Economics of Construction and Housing and Communal Services. 2022; 1(1):31-38. EDN IIIPWK. (rus.).

3. Kestelev N.A. The role of information modeling and automated technologies in the design and construction of high-rise buildings. Modern prospects for the development of flexible manufacturing systems in industrial civil engineering and the agro-industrial complex : collection of scientific articles of the 2nd All-Russian scientific and technical conference of young scientists, graduate students, masters and bachelors. 2024; 235-240. EDN QGFLSV. (rus.).

4. Lapidus A.A., Fedosov S.V., Petrukhin A.B., Kenewei E. Digital information modeling BIM is one of the possibilities of construction objects life cycle management. Construction Production. 2023; 4:32-36. DOI: 10.54950/265853402023432. EDN AWXPGA. (rus.).

5. Nikitin A.R. Evaluation of working documentation prepared using information modeling. Technologies, models and algorithms for modernizing science in modern geopolitical conditions : collection of articles from the National (All-Russian) scientific and practical conference with international participation. 2025; 62-67. EDN NNJCEG. (rus.).

6. Pridvizhkin S.V., Tyapochkin M.Yu., Startseva M.G. Estimation of construction costs in the Kortros group of companiesusing information model data analytics. Bulletin of SUSU. Series: Construction Engineering and Architecture. 2024; 24(4):41-49. DOI: 10.14529/build240406. EDN LCNAWB. (rus.).

7. Evsyagina V.A., Dmitrieva N.N. Verification of normative documents in the process of project documentation development with the help of an information model. Construction Economics. 2024; 4:160-164. EDN OACGCE. (rus.).

8. Gao S. A BIM-based ontological seismic multi-objective evaluation and optimisation design for buildings : PhD thesis. Cardiff University, 2025. URL: https://orca.cardiff.ac.uk/id/eprint/177611

9. Mandritsa P.M. Application of artificial intelligence in project documentation expertise: enhancing efficiency and quality in construction design. Innovation & Investment. 2024; 5:623-628. EDN SVSJLM. (rus.).

10. Hosseini M.A., Ravanshadnia M., Rahimzadegan M., Ramezani S. Next-Generation Building Condition Assessment: BIM and Neural Network Integration. Journal of Performance of Constructed Facilities. 2024; 38(6). DOI: 10.1061/JPCFEV.CFENG-4828

11. Ostadreza A., Shahhosseini V. Automated Construction Progress Monitoring Using Image Segmentation Trained on a Synthetic Dataset, 3D Reconstruction, and BIM. 2024. DOI: 10.2139/ssrn.5049036

12. Nikitin A.R., Sinenko S.A. Identification of indicators and dominant influencing the evaluation of project documentation. The Eurasian Scientific Journal. 2024; 16(5). EDN CWDXSG. (rus.).

13. Mishchenko D.S., Skurkan E.S. Features of the development of design and working documentation. Information systems and technologies of the agro-industrial complex and industrial civil engineering : collection of scientific articles of the 2nd International Scientific and Technical Conference. 2024; 279-283. EDN LUNCQJ. (rus.).

14. Redko V.S., Tsimbelman N.Y. Algorithm to analyze and control data quality and project quality using digital information model. Engineering journal of Don. 2024; 7(115):518-536. EDN ECDPWM. (rus.).

15. Rybakova A.O. Use of information models of modular elements at the stage of architectural and construction design of capital construction projects. Moscow, 2023; 201. EDN DHJXLQ. (rus.).

16. Tsay G.S., Staub-French Sh., Poirier E., Zadeh P., Pottinger R. BIM for FM: understanding information quality issues in terms of compliance with owner’s Building Information Modeling Requirements. Frontiers in Built Environment. 2023; 9. DOI: 10.3389/fbuil.2023.1117066. EDN YJKICO.

17. Candir E., Atasoy G. Exploring quality issues in building information models via structural design reviews. IOP Conference Series: Earth and Environmental Science. 2022; 1101(9):092011. DOI: 10.1088/1755-1315/1101/9/092011. EDN MVWWTE.

18. Gorodnova N.V., Lemeza V.A. Bim technolo-gies in the digital economy: world experience and Russian practice. Journal of Economics, Entrepreneurship and Law. 2022; 12(8):2241-2260. DOI: 10.18334/epp.12.8.115082. EDN XSISOS. (rus.).

19. Zhang C., Hu Z., Zhu H., Wang X., Gao J. Effects of silane on reaction process and microstructure of metakaolin-based geopolymer composites. Journal of Building Engineering. 2020; 32:101695. DOI: 10.1016/j.jobe.2020.101695

20. Khan A.A., Bello A.O., Arqam M., Ullah F. Integrating Building Information Modelling and Artificial Intelligence in construction projects: A review of challenges and mitigation strategies. Technologies. 2024; 12(10):185. DOI: 10.3390/technologies12100185

21. Nikitin A. Features of the use ai in generative design of building and structures. Journal of Mechanics of Continua and Mathematical Sciences. 2024; 19(5). DOI: 10.26782/jmcms.2024.05.00001

22. Liu H., Gan V.J.L., Cheng J.C.P., Zhou S.A. Automatic fine-grained BIM element classification using multi-modal deep learning (MMDL). Advanced Engineering Informatics. 2024; 61:102458. DOI: 10.1016/j.aei.2024.102458

23. Peng J., Liu X. Automated code compliance checking research based on BIM and knowledge graph. Scientific Reports. 2023; 13(1). DOI: 10.1038/s41598-023-34342-1

24. Kayhani N., McCabe B., Sankaran B. Semantic-aware quality assessment of building elements using graph neural networks. Automation in Construction. 2023; 155:105054. DOI: 10.1016/j.autcon.2023.105054

25. Madireddy S., Gao L., Din Z.U., Kim K., Senouci A., Han Z. et al. Large Language Model-Driven Code Compliance Checking in Building Information Modeling. Electronics. 2025; 14(11):2146. DOI: 10.3390/electronics14112146

26. Slepushkin D.V., Burlov D.Yu. Artificial intelligence and automation of design processes in construction: a bibliometric analysis. Vestnik MGSU [Monthly Journal on Construction and Architecture]. 2025; 20(3):440-455. DOI: 10.22227/1997-0935.2025.3.440-455. EDN MYHCAL. (rus.).

27. Candir E., Atasoy G. Exploring quality issues in building information models via structural design reviews. IOP Conference Series: Earth and Environmental Science. 2022; 1101(9):092011. DOI: 10.1088/1755-1315/1101/9/092011. EDN MVWWTE.

28. Sinenko S.A., Nikitin A.R. Determination of criteria for evaluating project documentation prepared using information modeling. Construction Production. 2024; 2:88-94. DOI: 10.54950/26585340_2024_2_88. EDN SGGPSM. (rus.).

29. Nikitin A.R., Sinenko S.A. On the results of the analysis of design documentation for construction, developed using information modeling technologies. Proceedings of Universities. Investment. Construction. Real estate. 2025; 15(2):(53):257-265. DOI: 10.21285/2227-2917-2025-2-257-265. EDN KPVVRM. (rus.).