رهپویه معماری و شهرسازی

رهپویه معماری و شهرسازی

گذار از بازنمایی به استنتاج: تبیین چارچوب مفهومی مدل‌سازی اطلاعات ساختمان شناختی در عصر هوش مصنوعی

نوع مقاله : مقاله پژوهشی

نویسندگان
1 کارشناس ‌ارشد معماری، گروه معماری، دانشکده معماری و شهرسازی، دانشگاه علم و صنعت، تهران، ایران.
2 دانشیار گروه معماری، دانشکده هنر و معماری، دانشگاه پیام نور، تهران، ایران.
چکیده
تحولات اخیر در هوش مصنوعی، یادگیری ماشین، طراحی مولد و مدل‎های زبانی بزرگ، ظرفیت‎های تازه‎ای برای بازتعریف نقش مدل‎سازی اطلاعات ساختمان در صنعت معماری، مهندسی، ساخت و بهره‎برداری فراهم کرده است. در حالی که نسل‎های اولیه بیم بر بازنمایی دیجیتال و مستندسازی متمرکز بودند، روندهای نوظهور نشان می‎دهد این فناوری در حال حرکت به سوی نقش‎های تحلیلی، استنتاجی و تصمیم‎یارانه است. با وجود گسترش مطالعات مرتبط، هنوز چارچوب مفهومی جامعی برای تبیین این گذار و تحلیل پیامدهای آن محدود است.
پژوهش حاضر یک مطالعه نظری ـ مفهومی مبتنی بر تحلیل انتقادی اسناد و ادبیات علمی منتشرشده در بازه زمانی 2019 تا 2025 است که با رویکرد توصیفی ـ تحلیلی روند همگرایی مدل‎سازی اطلاعات ساختمان و هوش مصنوعی را بررسی می‎کند. یافته‎ها نشان می‎دهد تحول بیم را می‎توان در قالب مدلی سه‎نسلی شامل بیم بازنمایی‎محور، یکپارچه‎سازی‎محور و استنتاج‎محور صورت‎بندی کرد. این گذار در چهار مسیر اصلی شامل تحلیل پیش‎بین عملکردی، طراحی مولد، هوشمندسازی پردازش اطلاعات و تعامل شناختی انسان ـ ماشین رخ می‎دهد. در این میان، شکاف میان توسعه فناورانه و بلوغ سازمانی، ضعف قابلیت همکاری میان سامانه‎ها و کمبود شایستگی‎های ترکیبی حرفه‎ای از مهم‎ترین موانع این تحول هستند. نتایج نشان می‎دهد آینده بیم در شکل‎گیری همکاری شناختی مؤثر میان انسان و ماشین تحقق می‎یابد؛ به طوری که بیم به عنوان حافظه ساختاریافته طرح، هوش مصنوعی به مثابۀ موتور تحلیل، و معمار در مقام مرجع نهایی قضاوت فضایی و ارزشی عمل می‎کند.
کلیدواژه‌ها
موضوعات

عنوان مقاله English

From Representation to Inference: Conceptualizing Cognitive BIM in the Age of Artificial Intelligence

نویسندگان English

Amir Hossein Karimi 1
Heidar Jahanbakhsh 2
1 Master in architecture، Department of architecture, Faculty of Architecture and Urban Planning, Iran University of Science and Technology (IUST), Tehran, Iran.
2 Associate Professor of Architecture Department, Faculty of Art and Architecture, Payame Noor University, Tehran, Iran.
چکیده English

Introduction
The Architecture, Engineering, Construction, and Operations (AECO) industry is currently undergoing a profound digital transformation, fundamentally altering traditional design, construction, and lifecycle management processes. Among the technological infrastructures driving this transformation, Building Information Modeling (BIM) has emerged as one of the most influential paradigms for integrating project information, improving interdisciplinary coordination, and enabling lifecycle-based decision-making.
Since its early adoption, BIM has been primarily understood as a digital representation framework for organizing, documenting, and exchanging project-related information. Its core value has historically resided in improving geometric modeling accuracy, facilitating interdisciplinary collaboration, minimizing conflicts, and supporting lifecycle information management. However, despite these advantages, BIM has often remained confined to operational applications focused on visualization, documentation, and data management rather than serving as an active analytical or decision-support environment.
Recent developments in artificial intelligence (AI), including machine learning, generative design, large language models, and cognitive computing systems, have significantly expanded the functional potential of BIM environments. The integration of these technologies has introduced new capabilities such as predictive performance analysis, automated design generation, intelligent information processing, and conversational interaction between users and digital systems. These developments suggest that BIM is gradually evolving from a passive information repository into an active inferential and cognitive system capable of supporting design reasoning and strategic decision-making.
Despite growing research on AI-driven BIM applications, existing studies remain fragmented and largely focused on technical implementations rather than conceptual interpretation. There remains a significant lack of comprehensive theoretical frameworks capable of explaining this transformation as a broader epistemological and professional shift within architectural practice.
This study addresses this gap by proposing a conceptual model for understanding BIM evolution as a three-generational transition from representational BIM to inferential and cognitive BIM. It further examines the technological, organizational, and professional implications of this transformation and discusses its consequences for the future role of architects in AI-augmented design environments.
Research Method
This study adopts a conceptual research approach grounded in critical document analysis and descriptive-analytical interpretation. Relevant scholarly literature published between 2019 and 2025 was systematically identified through targeted searches in major academic databases focusing on BIM evolution, artificial intelligence integration, digital transformation, and intelligent systems in the AECO sector.
The selected literature was evaluated based on three criteria: conceptual relevance to the research problem, scientific credibility and citation impact, and recency of publication. Through purposive sampling, studies representing major developments in AI-BIM convergence were selected for conceptual synthesis.
The analytical process involved thematic extraction and comparative interpretation. Key themes included BIM functional transformation, AI-enhanced analytical capabilities, human-system interaction models, interoperability challenges, organizational maturity, and professional competency requirements.
Through iterative comparative analysis, recurring patterns of BIM evolution were identified and synthesized into a conceptual framework consisting of three distinct developmental generations: representational BIM, integration-oriented BIM, and inferential BIM (Cognitive BIM). The resulting conceptual framework was subsequently used to interpret the broader theoretical and professional implications of BIM transformation for architectural design practice.
Findings
The findings indicate that BIM evolution can be understood as a paradigmatic transition across three distinct generations.
The first generation, representational BIM, focuses primarily on geometric modeling, documentation, clash detection, and structured information storage. At this stage, BIM functions as a passive digital infrastructure for organizing project information and improving technical coordination.
The second generation, integration-oriented BIM, extends beyond representation by enabling interdisciplinary collaboration, lifecycle data management, and structured interoperability across professional domains. BIM becomes a collaborative platform facilitating coordinated information exchange across project phases.
The third generation, inferential BIM or Cognitive BIM, introduces analytical reasoning, predictive intelligence, automated suggestions, and adaptive learning capabilities. Here, BIM evolves into an intelligent decision-support system capable of interpreting patterns, generating alternatives, and participating in design reasoning processes.
This transformation is manifested across four primary developmental pathways:
Predictive performance analysis enables BIM systems to forecast energy behavior, environmental performance, and operational outcomes through machine learning models trained on structured project data.
Generative design integration allows automated production and evaluation of multiple design alternatives based on predefined performance constraints and qualitative objectives.
Intelligent information processing improves automated classification, semantic enrichment, and interoperability optimization across fragmented digital ecosystems.
Cognitive human-machine interaction introduces conversational BIM environments through large language models, allowing natural-language communication, reasoning-based feedback, and accessible decision support for both experts and non-specialist stakeholders.
However, the study also reveals substantial barriers to full Cognitive BIM realization.
A critical challenge lies in the gap between technological capability and organizational maturity. Many organizations remain limited to tool-level BIM adoption and lack institutional structures necessary for strategic AI integration.
Interoperability limitations continue to constrain seamless data exchange between heterogeneous systems, reducing the effectiveness of AI-driven reasoning processes.
Professional competency gaps also remain significant. Effective use of Cognitive BIM requires hybrid expertise combining technical literacy, strategic management capability, critical reasoning, and algorithmic understanding.
Institutional resistance to organizational transformation further slows BIM evolution, particularly in environments lacking adaptive policy frameworks and innovation-oriented educational infrastructures.
Discussion
The conceptual findings suggest that BIM evolution should not be understood merely as technical software advancement but as a broader reconfiguration of architectural cognition and professional agency.
Traditional BIM environments support data representation and procedural coordination; Cognitive BIM introduces analytical participation within the design process itself. This marks a shift from digitization toward intelligent inference.
Such transformation fundamentally redefines the architect’s professional role. Rather than acting solely as model producer and coordinator, the architect becomes a cognitive orchestrator responsible for evaluating, interpreting, and strategically guiding AI-generated design propositions.
This transition does not imply replacement of human creativity. Instead, it establishes a collaborative relationship in which AI augments analytical capacity while human judgment retains authority over spatial quality, cultural meaning, ethical interpretation, and value-based decision-making.
The study therefore proposes understanding Cognitive BIM as a socio-cognitive system emerging through interaction among structured data, intelligent algorithms, organizational ecosystems, and human critical judgment.
This perspective challenges reductionist interpretations that frame BIM intelligence purely as computational sophistication. Instead, BIM intelligence should be evaluated by its capacity to support collective reasoning, interpretive dialogue, and augmented design cognition.
Educationally, this transformation necessitates moving beyond software-centered instruction toward cultivating digital cognitive literacy. Future architects must develop competencies in algorithmic reasoning, critical interpretation of machine-generated outputs, and collaborative human-AI decision-making.
Conclusion
This study conceptualizes BIM evolution as a three-generational transition from representational modeling to inferential cognitive systems. The proposed framework demonstrates that the future of BIM lies not in increasingly complex modeling tools but in enabling effective cognitive collaboration between humans and intelligent systems. In this emerging paradigm, BIM functions as the structured memory of architectural projects, AI serves as the analytical and inferential engine, and architects remain the ultimate evaluators of spatial and cultural value.
The primary contribution of this research lies in introducing a conceptual framework that interprets BIM transformation as a cognitive-organizational reconfiguration rather than a purely technical development. This perspective highlights that successful implementation depends not only on computational advancement, but also on institutional maturity, interdisciplinary competencies, and the capacity of architectural practice to critically integrate intelligent systems into decision-making processes.
Future research should focus on operationalizing Cognitive BIM implementation frameworks, improving explainability and transparency of AI-driven architectural systems, and redesigning architectural education around critical human-AI interaction competencies. Ultimately, the next generation of intelligent architectural design will depend not on competition between humans and machines, but on the productive synthesis of human interpretive intelligence and computational inference.

کلیدواژه‌ها English

Building Information Modeling (BIM), Cognitive BIM, Artificial Intelligence, Architectural Design, Digital Transformation, Human–Machine Interaction
-        اسلامی، سیدغلامرضا.، و اسلامی، سید یحیی (1403). مدل سازی مفهومی و معماری اندیشه. تهران: دانشگاه تهران.
Ajtayné Károlyfi, K., & Szép, J. (2023). A Parametric BIM Framework to Conceptual Structural Design for Assessing the Embodied Environmental Impact. Sustainability, 15(15), 11990. https://doi.org/10.3390/su151511990
Alves, J. L., Palha, R. P., & de Almeida Filho, A. T. (2025). Towards an integrative framework for BIM and artificial intelligence capabilities in smart architecture, engineering, construction, and operations projects. Automation in Construction, 174, 106168. https://doi.org/10.1016/j.autcon.2025.106168
Awe, M., Malhi, A., Budka, M., Mavengere, N., & Dave, B. (2025). Towards 4D BIM: A systematic literature review on challenges, strategies and tools in leveraging AI with BIM. Buildings, 15(7), 1072. https://doi.org/10.3390/buildings15071072
Bagasi, O., Nawari, N. O., & Alsaffar, A. (2025). BIM and AI in Early Design Stage: Advancing Architect–Client Communication. Buildings, 15(12), 1977. https://doi.org/10.3390/buildings15121977
Cascone, S., Parisi, G., & Caponetto, R. (2024). BIM-Based Strategies for the Revitalization and Automated Management of Buildings: A Case Study. Sustainability, 16(16), 6720. https://doi.org/10.3390/su16166720
Chen, Y., Yang, Y., Luo, L., & Ding, C. (2023). Impacts of BIM policy on the technological progress in the Architecture, Engineering, and Construction (AEC) industry: evidence from China. Journal of Asian Architecture and Building Engineering, 23(6), 1997–2011. https://doi.org/10.1080/13467581.2023.2278888
Correa, J. G. A., Alves, J. L., Homrich, A. S., & de Carvalho, M. M. (2025). Evolving skillsets of architecture, engineering and construction sector: unveiling the interplay between project management, BIM, strategic and operational skills. Engineering, Construction and Architectural Management. https://doi.org/10.1108/ECAM-05-2024-0670
Cui, Q., Hu, X., Liu, X., Zhao, L., & Wang, G. (2021). Understanding Architectural Designers’ Continuous Use Intention Regarding BIM Technology: A China Case. Buildings, 11(10), 448. https://doi.org/10.3390/buildings11100448
Das, K., Khursheed, S., & Paul, V. K. (2025). The impact of BIM on project time and cost: insights from case studies. Discover Materials, 5, 25. https://doi.org/10.1007/s43939-025-00200-2
Dong, Y., Zhan, Z., Hu, Y., Doe, D. M., & Han, Z. (2025). AI BIM coordinator for non-expert interaction in building design using LLM-driven multi-agent systems. Automation in Construction, 180, 106563. https://doi.org/10.1016/j.autcon.2025.106563
Feng, H., Kassem, M., Greenwood, D., & Doukari, O. (2023). Whole building life cycle assessment at the design stage: a BIM-based framework using environmental product declaration. International Journal of Building Pathology and Adaptation, 41(1), 109–142. https://doi.org/10.1108/IJBPA-06-2021-0091
Gartoumi, K. I., Zaki, S., & Aboussaleh, M. (2023). BIM interoperability for architecture and engineering: A case study. Materials Today: Proceedings. https://doi.org/10.1016/j.matpr.2023.05.408
Gath-Morad, M., Aguilar Melgar, L. E., Conroy-Dalton, R., & Hölscher, C. (2022). Beyond the shortest-path: Towards cognitive occupancy modeling in BIM. Automation in Construction, 135, 104131. https://doi.org/10.1016/j.autcon.2022.104131
Hajirasouli, A., Banihashemi, S., Sanders, P., & Rahimian, F. (2024). BIM-enabled virtual reality-based pedagogical framework in architectural design studios. Smart and Sustainable Built Environment, 13(6), 1490–1510. https://doi.org/10.1108/SASBE-07-2022-0149
Heaton, J., Parlikad, A. K., & Schooling, J. (2019). Design and development of BIM models to support operations and maintenance. Computers in Industry, 111, 172–186. https://doi.org/10.1016/j.compind.2019.08.001
Jasiński, A. (2020). Impact of BIM implementation on architectural practice. Architectural Engineering and Design Management, 17(5–6), 447–457. https://doi.org/10.1080/17452007.2020.1854651
Jiao, Y., & Cao, P. (2023). Research on optimization of project design management process based on BIM. Buildings, 13(9), 2139. https://doi.org/10.3390/buildings13092139
Kutá, D., & Faltejsek, M. (2025). The role of artificial intelligence in the transformation of the BIM environment: Current state and future trends. Applied Sciences, 15(18), 9956. https://doi.org/10.3390/app15189956
Lidelöw, S., Engström, S., & Samuelson, O. (2023). The promise of BIM? Searching for realized benefits in AECO industries. Journal of Building Engineering, 76, 107067. https://doi.org/10.1016/j.jobe.2023.107067
Li, L., Yuan, J., Tang, M. et al. (2021). BIM-enabled building lifecycle management system. Automation in Construction, 129, 103814. https://doi.org/10.1016/j.autcon.2021.103814
Ma, W., Wang, X., Wang, J. et al. (2021). Generative design in BIM: approaches and requirements. Sensors, 21(16), 5439. https://doi.org/10.3390/s21165439
Martin, H., Watson, C., & Brooks, T. (2024). Synergizing BIM and RIBA in architectural practice. Architectural Engineering and Design Management, 21(1), 35–59. https://doi.org/10.1080/17452007.2024.2361284
Park, J. J., Kim, K., Ji, S. -Y., & Jun, H. J. (2024). BIM-based repair history management for architectural heritage. Applied Sciences, 14(6), 2315. https://doi.org/10.3390/app14062315
Pavard, A., de Paula, N., & Poirier, E. (2025). Organizational competencies for BIM adoption. Construction Innovation. https://doi.org/10.1108/CI-03-2024-0092
Rupasinghe, C., Ranadewa, K. A. T. O., Jayanetti, J. K. D. D. T., & Perera, B. A. K. S. (2025). Green-BIM team framework for sustainable construction. Engineering, Construction and Architectural Management. https://doi.org/10.1108/ECAM-03-2024-0276
Sampaio, A. Z., Gomes, A. M., Sánchez-Lite, A. et al. (2023). BIM methodology in structural design collaboration. Buildings, 13(1), 31. https://doi.org/10.3390/buildings13010031
Tsikas, P., Chassiakos, A., Papadimitropoulos, V., & Papamanolis, A. (2025). BIM-based machine learning for energy performance. Energies, 18(1), 201. https://doi.org/10.3390/en18010201
Tran, T. B., Dinh, T. P., Nguyen, T. D. H. N. et al. (2025). BIM-based automation for vernacular architecture design. Applied Sciences, 15(12), 6776. https://doi.org/10.3390/app15126776
Utkucu, D., Ying, H., Wang, Z., & Sacks, R. (2024). BIM object classification using machine learning. Advanced Engineering Informatics, 61, 102503. https://doi.org/10.1016/j.aei.2024.102503
Xiang, Y., Mahamadu, A. -M., & Florez-Perez, L. (2024). BIM applicability across design stages. Journal of Building Engineering, 95, 110030. https://doi.org/10.1016/j.jobe.2024.110030
Valdebenito, R., & Forcael, E. (2025). Integrating artificial intelligence and BIM in construction: Systematic review and quantitative comparative analysis. Applied Sciences, 15(23), 12470. https://doi.org/10.3390/app152312470
Yilmaz, G., Akcamete, A., & Demirors, O. (2023). BIM capability assessment in AECO industry. Computers in Industry, 147, 103861. https://doi.org/10.1016/j.compind.2023.103861
Zhang, S., Tang, Y., Zou, Y., & Li, M. (2024). BIM and PLM integration for design optimization. Scientific Reports, 14, 26153. https://doi.org/10.1038/s41598-024-75940-x

  • تاریخ دریافت 30 اردیبهشت 1405
  • تاریخ بازنگری 20 تیر 1405
  • تاریخ پذیرش 25 تیر 1405