Ageing bridge stocks and rising traffic loads in Latin America and worldwide demand cost-effective maintenance strategies. Building Information Modelling (BIM) and its convergence with digital-twin, IoT and AI techniques have shown promise, yet their adoption for bridge upkeep remains fragmentary. This review aimed to (i) synthesise current scientific evidence on BIM-based bridge maintenance, (ii) classify methodologies and tools through a domain taxonomy, and (iii) identify research gaps that hinder large-scale implementation. A PRISMA-guided systematic literature review was conducted in Scopus and Web of Science (search cut-off = 1 February 2024). Inclusion criteria targeted peer-reviewed, open-access studies (2020–2024) that applied BIM to the maintenance of existing bridges. Twenty-five articles met the criteria and were appraised with Mixed-Methods Appraisal Tool (MMAT 2018). Seven dominant research themes were identified, with damage visualization (7 studies) and 3-D geometric modelling (6) being the most frequent, followed by information exchange/management (4). Specifically, LiDAR and photogrammetry enabled sub-centimetre models; Convolutional Neural Networks (CNN) and You Only Look Once (YOLO) algorithms reached mean average precision up to 0.91 for crack detection. Digital-twin workflows reduced operating costs while requiring higher upfront investment. A seven-domain taxonomy and a cost–technology comparison table is proposed. Key barriers reported include IFC 4.3 interoperability, high LiDAR costs ( > 10% of annual budgets), limited visual-programming skills, and cybersecurity concerns in cloud-IoT integrations. BIM supports preventive, data-driven bridge maintenance and has been linked to lower operating costs in several studies; mainstream adoption requires IFC 4.3 based interoperability, targeted training, and open-standard workflows. Future research should focus on standardised performance metrics, edge-AI monitoring and blockchain-secured data exchange.
Medrano-Sanchez et al. (2025) studied this question.