This thesis explores the potential of utilizing Generative AI to enhance knowledge contextualization among stakeholders to promote circularity in the construction industry. Circularity, essential for sustainable development, demands a shift from the linear “take- make-dispose” model to a regenerative approach that minimizes resource consumption and waste. The construction industry, a significant contributor to global material use and waste, faces unique challenges in implementing circularity due to its technical complexity and the wide range of stakeholders involved. Consultants, who play a crucial role in bridging the gap between technical knowledge and practical application, are central to overcoming these challenges. This study focuses on consultants and their ability to disseminate and contextualize knowledge, a process hindered by technical language, the varying expertise of stakeholders, and the multi-disciplinary nature of construction projects.
The research investigates the current applications of Generative AI, with a particular focus on ChatGPT, as a tool for knowledge contextualization in circular construction projects. Through a comprehensive literature review and qualitative data collection from semi-structured interviews with industry professionals, this thesis identifies both the barriers to effective knowledge-sharing stakeholders encounter in circular projects and the potential affordances of ChatGPT in overcoming these challenges. The state-of-the-art capabilities of ChatGPT include features such as contextual comprehension, few-shot learning, and coherent content generation, which allow it to process and present complex information in a user-friendly format tailored to the specific needs of stakeholders. Its affordances, such as knowledge retrieval, organization, and tailoring, enable ChatGPT as a powerful tool to create customized knowledge outputs, making it particularly suited for addressing the diverse and technical demands of the construction industry. The analysis of the interviews highlights that the limited understanding of circularity principles among stakeholders stems from the complexity of technical documentation, the information overload of the documents regarding circularity, and the lack of tailored information that adapts to the specific needs of different actors and projects in the construction process.
The thesis investigates the possibility of ChatGPT as a solution to these issues by dynamically tailoring technical information to match the expertise and tasks of diverse stakeholders. The research studies how ChatGPT's features and affordances in knowledge management can enhance the contextualization of circularity knowledge, improving the accessibility and usability of complex information for non-experts. This approach fosters better collaboration among stakeholders and accelerates the adoption of circularity practices, often slowed by misunderstandings and technical jargon.
The findings suggest that while the potential of ChatGPT in knowledge management is vast, its implementation is still in the early stages. The study proposes a framework for integrating Generative AI into the knowledge-sharing processes of the circular construction industry, aiming to optimize knowledge dissemination and improve stakeholder comprehension. The framework is based on tailoring the capabilities of AI to support consultants in their pivotal role of contextualizing knowledge. By addressing the identified barriers to the adoption of this tool, this thesis contributes to the ongoing discourse on sustainable construction practices and highlights the importance of technology in advancing the industry towards circularity.
This thesis concludes by offering practical guidelines for the integration of Generative AI in circular construction projects, underscoring the need for further research into its long-term implications. The proposed guidelines for integrating ChatGPT into the construction industry focus on three key phases: preparation, where stakeholders are trained to understand the AI’s capabilities; pilot testing, where AI-assisted knowledge contextualization is trialed in a controlled setting; and deployment, where the AI is fully integrated into project workflows with ongoing monitoring to ensure its effectiveness in enhancing communication and knowledge sharing. The research lays the groundwork for future studies aimed at refining AI applications in this context and ensuring that the construction industry can meet its sustainability goals through improved knowledge management practices.