New paper: Prioritising generative AI adoption challenges in construction projects: a Fuzzy Analytic Hierarchy Process approach
Generative AI is increasingly viewed as a technology to help address the construction industry's long-standing productivity challenges. It is therefore important to examine how these tools can be successfully adopted in construction projects. This study reveals that the most critical adoption challenges are not technological limitations, but concerns related to governance, data quality, and organisational readiness. The findings highlight the need to look beyond technology itself and consider the broader conditions required for successful implementation. These are the key takeaways from research by Alejandro Vidal Fernández (one of our former MSc students) and Dr Yixue Shen.
Moving from identifying challenges to prioritising them: What practitioners should know
While existing studies have explored generative AI opportunities and challenges in construction, much of this work remains conceptual or examines challenges in isolation. This study shifts the focus from identifying adoption challenges to understanding their relative importance in practice, i.e which challenges should be prioritised when planning generative AI implementation strategies for practitioners and organisations.
What the study did
- The authors conducted a systematic literature review to identify the key challenges associated with generative AI adoption in construction projects. The identified challenges were grouped into four categories: domain-specific, technological, adoption-related, and ethical challenges.
- Based on the literature review findings, an expert questionnaire was developed and completed by construction professionals from a range of industry backgrounds and experience levels. The responses were analysed using the Fuzzy Analytic Hierarchy Process (FAHP) approach with sensitivity analysis, enabling the relative importance of each challenge to be assessed and prioritised reliably according to practitioners’
What they found
- Data privacy, security, and legal liability emerged as the most critical adoption challenge, highlighting the need for robust legal, ethical, and data governance frameworks.
- Interestingly, technological challenges were perceived as comparatively less critical and more manageable than organisational and contextual challenges.
- The findings also reveal a misalignment between general-purpose generative AI tools and the specialised requirements of construction projects, emphasising the need for using generative AI tools with stronger construction-specific knowledge, higher-quality domain datasets, and workforce capabilities.
Why this matters for practice
For practitioners seeking to implement generative AI in construction projects, the findings suggest that:
- Organisations should prioritise the development of robust ethical, legal, and governance frameworks to support the responsible adoption of generative AI.
- Investment in high-quality construction-specific datasets and workforce upskilling initiatives is essential to maximise the effectiveness of generative AI adoption.
- Generative AI solutions should be aligned with construction-specific contexts, standards, and requirements rather than relying solely on general-purpose models.
- Successful generative AI adoption requires a holistic approach that considers technical, organisational, contextual, and ethical challenges, as their relative importance may vary across implementation contexts.
Together, these insights support a more structured and evidence-based approach to generative AI adoption in construction projects.
If you would like to find out more about this research please access the full paper here or contact Dr Yixue Shen ([email protected]).
Reference
Vidal, A. and Shen, Y. 2026. Prioritising generative AI adoption challenges in construction projects: a Fuzzy Analytic Hierarchy Process approach. International Journal of Construction Management. [Online]. pp.1-16. https://doi.org/10.1080/15623599.2026.2664476