starting point: 

How is design knowledge embedded in Architecture, Engineering, and Construction (AEC) sector data (BIM, CAD, etc.) and what methods (for example, artificial intelligence, machine learning) are available for harvesting design data embedded in past projects?

This topic may include consideration of: the usefulness of design data and its security; along with legal, ethical, and commercial challenges to using design data.

project summary: 

1. Introduction: The Challenge of Lost Knowledge in Architecture 

The Architecture, Engineering, and Construction (AEC) sector is undergoing a significant digital transformation, driven by demands for higher standards in safety, cost, and efficiency. However, the industry’s fragmented, project-based nature and reactive approach to innovation potentially hinder its ability to leverage digital tools for organisational learning (Criado-Perez et al., 2022). Architectural design processes generate vast amounts of valuable knowledge, yet this knowledge is often ephemeral, either trapped in the memories of team members or invisibly embedded within complex computer models. Consequently, architectural firms struggle with systemic knowledge loss, leading to instances of “reinventing the wheel,” decreased project efficiency, and missed opportunities for innovation (Carrillo et al., 2000). 

This research addresses a critical problem: the inability of architectural organisations to systematically learn from historical project data to improve future decision-making. This challenge is rooted in the ephemeral nature of project knowledge, which is often tacit, distributed amongst team members, and ultimately lost when teams disband. This problem is exacerbated by ineffective or inadequately resourced post-project reviews (PPRs) that are unable to capture knowledge in a reusable format, and by the competitive, multi-organisational structure of project teams, which can create barriers to knowledge sharing (Kamara et al., 2003). 

2. Limitations of Current Approaches: Beyond the “BIM Utopia” 

Recent efforts to solve this problem have focused on technology-driven solutions, particularly Building Information Modelling (BIM), as a unified data platform. While beneficial for data interoperability, these approaches often fall short of enabling process interoperability. They are guided by a “BIM utopia” vision that overlooks the nonlinear, sociotechnical reality of design, where informal communication, negotiation, and ad-hoc problem-solving are critical to success (Miettinen & Paavola, 2014). This oversight points to two fundamental knowledge gaps in existing research: 

Knowledge Gap 1: Past research has been unable to show how knowledge capture in the architectural design process can move beyond static, top-down models to consider the complex, contextual, and evolutionary nature of design knowledge. 

Knowledge Gap 2: Past research has identified the importance of both analysing data assets and design process dynamics, but the implications and possible applications of their combination are largely unexplored.  

3. A Sociotechnical Computational Framework for Project Knowledge Capture and Reuse 

This research aims to develop a sociotechnically-informed computational framework that captures and makes accessible the rich process knowledge generated during design. The goal is to enable firms to achieve ‘double-loop learning’, moving beyond simple error correction to fundamentally question and modify their guiding standards and processes based on historical data (Argyris & Schön, 1996). 

The conceptual foundation of this work rests on two pillars: 

• Sociotechnical Systems (STS) Theory: Viewing the design process as a complex dynamic system of people, ideas, decisions, and technical artefacts, where outcomes depend on the interplay between social interactions and technical processes (Pirzadeh et al., 2021). 

• A Bottom-Up View of Knowledge: Adopting a “no-model” philosophy that allows knowledge structures to emerge organically from the analysis of practitioners’ communications and the artefacts they create, rather than imposing rigid, predefined schemas (El-Diraby, 2023). 

The research is structured in three major stages. First, the project begins by developing an understanding of the architectural design process through a Systematic Literature Review (SLR) and exploratory case studies of contemporary Australian architectural projects (2015-2025), which together will identify and define its fundamental sociotechnical patterns and components. Following this, the second stage critically assesses existing methods for design knowledge capture, such as BIM-centric models and formal process maps, evaluating them against the patterns identified in the initial stage to determine their principal limitations. This provides a gap analysis that informs the development of the proposed computational framework. Finally, the third stage focuses on developing and validating the proposed mathematical and computational framework designed to model the design process as an evolving sociotechnical network. This data-driven model, which is envisioned to leverage a temporal graph structure to map the dynamic interactions between actors, artefacts, and decisions, will be validated through retrospective analysis of completed projects and evaluated by a panel of industry and academic experts. 

4. Anticipated Impact and Contributions 

This research is expected to deliver both theoretical and practical contributions. On a theoretical level, it will offer a new, more nuanced model for understanding architectural design as a complex sociotechnical system. On a practical level, the framework will provide the foundations for a structured methodology for architectural firms to learn from their project knowledge. 

By making the tacit knowledge embedded in design processes explicit and reusable, the findings developed in this research address the need to: 

• Mitigate systemic knowledge loss as teams transition between projects. 

• Support more evidence-based decision-making, particularly for less experienced staff. 

• Foster a culture of continuous organisational learning and proactive innovation. 

Ultimately, this research seeks to help transform how architectural project knowledge is captured and applied, enhancing operational efficiency and improving the quality of design outcomes. 

Key References 

Argyris, C., & Schön, D. (1996). Organizational Learning II: Theory, Method and Practice. Reading, MA: Addison-Wesley.  

Carrillo, P. M., Anumba, C. J., & Kamara, D. J. (2000). Knowledge Management Strategy for Construction: Key I.T. and Contextual Issues. Proceedings of the Inter. Conf. on Construction IT, 28-30 June, Reykjavik, Iceland, Icelandic Building Research Institute, 155-165. 

Criado-Perez, C., Shinkle, G. A., Höllerer, M. A., Sharma, A., Collins, C., Gardner, N., Hank Haeusler, M., & Pan, S. (2022). Digital Transformation in the Australian AEC Industry: Prevailing Issues and Prospective Leadership Thinking. Journal of Construction Engineering and Management, 148(1), 05021012. https://doi.org/10.1061/(ASCE)CO.1943-7862.0002214 

El-Diraby, T. E. (2023). How typical is your project? The need for a no-model approach for information management in AEC. Journal of Information Technology in Construction, 28, 19–38. https://doi.org/10.36680/j.itcon.2023.002 

Kamara, J. M., Anumba, C. J., & Carrillo, P. M. (2003). Conceptual framework for live capture and reuse of project knowledge. Proc., CIB W78’s 20th Int. Conf. on Information Technology for Construction, CIB, Rotterdam, Netherlands, 178–185. 

Miettinen, R., & Paavola, S. (2014). Beyond the BIM utopia: Approaches to the development and implementation of building information modeling. Automation in Construction, 43, 84–91. https://doi.org/10.1016/j.autcon.2014.03.009 

Pirzadeh, P., Lingard, H., & Blismas, N. (2021). Design Decisions and Interactions: A Sociotechnical Network Perspective. Journal of Construction Engineering and Management, 147(10), 04021110. https://doi.org/10.1061/(ASCE)CO.1943-7862.0002136