Artificial Intelligence Based Estimation of Offsite Panel Requirements in Building Construction: Challenges and Opportunities

Résumé fourni par la source

The construction industry is increasingly adopting offsite panelised systems, particularly light gauge steel (LGS) panel solutions, as a means of improving productivity, reducing waste, and accelerating project delivery. However, a critical challenge remains in accurately determining the quantity of LGS panels and any supporting hot-rolled steel (HRS) requirements for a given building project in the early stages of tendering, design, and planning. Estimating LGS and HRS quantities involves complex interdependencies between architectural configurations, structural requirements, building typologies, and site-specific constraints. Traditional estimation methods, which rely heavily on expert judgement and manual quantification processes, are time-consuming, inconsistent, and prone to error, often resulting in procurement inefficiencies. Despite growing research on artificial intelligence (AI) and machine learning (ML) applications in construction, their specific application to offsite LGS panel quantity estimation remains largely unexplored. Data-driven approaches capable of processing multiple building parameters simultaneously are needed to generate reliable quantity predictions for early-stage decision-making. The presented research proposes a conceptual framework for applying AI and ML techniques to estimate LGS panel and HRS requirements for building projects. The framework identifies key input parameters including building geometry, floor area, storey height, structural system type, complexity, and opening ratios. A structured methodology is outlined encompassing data requirements, feature identification, model selection considerations, and validation approaches. The framework addresses implementation challenges including data scarcity in offsite construction, workflow integration, and interpretability of model outputs for industry practitioners. The proposed framework has the potential to improve the accuracy and efficiency of LGS and HRS quantity estimation, support better procurement planning, and reduce material wastage. This conceptual contribution, developed through a University College Dublin – Evolusion Innovation partnership, provides a structured foundation for future empirical studies and AI tool development in offsite construction.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.