Error Prediction in Industrialized Construction: A Framework for AI-Powered Error Prediction in the On-Site phase of IHB

Abstract

ABSTRACT The construction sector faces notable productivity challenges, as it is known for having one of the lowest productivity levels compared to other industries. Industrialized House Building (IHB) has emerged as a solution for low productivity. However, the sector still faces productivity challenges as delays and cost overruns still exist, mainly due to errors and variations in IHB projects, especially in the on-site stage. To address the challenge of errors emerging in on-site IHB projects, this study aims to investigate the possibility of implementing AI tools for error prediction to mitigate errors in the on-site stage of IHB projects. The study employs an abductive approach through qualitative data collected from a literature review, site visits, interviews, and inspection report analysis in a thematic approach. The study identifies common errors and their impacts, emphasizing the importance of early intervention and predictive technologies in mitigating errors. Key findings reveal that AI can be employed for error prediction, enhancing resource allocation and planning, and minimizing projects’ rework. The study proposes a conceptual framework for an AI error prediction tool in the on-site stage of IHB projects to assist project managers in the decision-making process. This study contributes to IHB’s error management practices, bridging the gap between the on-site stage of IHB and AI, and serving as a roadmap for IHB companies to implement AI tools effectively. Finally, the research highlights the necessity of robust data management practices and continuous improvement to leverage AI’s potential in the IHB project.