Artificial intelligence in masonry engineering: Comprehensive review of design, analysis, and maintenance applications

Masonry is one of the most widely used construction systems worldwide, but its heterogeneity, anisotropy, and nonlinear behaviour continue to challenge conventional design, assessment, and maintenance approaches. Recent advances in artificial intelligence (AI) and machine learning (ML) offer powerful data-driven tools to address these challenges. This paper presents a critical review of AI and ML applications in masonry engineering, synthesising research across material, element, and structural scales, with emphasis on studies published between 2023 and 2025. The literature is organised into four domains: prediction of masonry behaviour, AI-assisted structural assessment, AI-enabled structural health monitoring, and AI for design and optimisation. Findings show that supervised and ensemble models achieve high accuracy in predicting mechanical properties and capturing nonlinear interactions beyond empirical formulations. Hybrid AI–physics frameworks support seismic vulnerability assessment, digital twins, and rapid evaluation of existing and heritage masonry structures. Advances in computer vision enable scalable, non-destructive monitoring, while optimisation and generative AI improve performance, cost, and sustainability trade-offs. Persistent challenges include data quality, model generalisability, validation, and standardisation. The review concludes that AI should augment, not replace, engineering judgement and provides a roadmap for responsible adoption. • Comprehensive review of AI/ML applications in masonry engineering. • Four key domains are identified: strength prediction, structural modelling, SHM, design optimisation. • AI models outperform empirical and FE methods in masonry behaviour prediction. • Deep learning and computer vision for automated crack, spalling, condition assessment. • Hybrid AI–traditional workflow, future needs in standardisation and validation.