Algorithmic workflow for digitising inspection data in railway tunnel maintenance

Abstract

Purpose This study addresses the lack of structured, data-driven approaches to infrastructure asset management, demonstrating how integrating building information modelling (BIM) and digital twin methodologies can enhance the maintenance and lifecycle management of railway tunnels.Design/methodology/approach A structured data-to-BIM workflow was co-developed with Infraestruturas de Portugal (IP), using inspection and monitoring data from 79 railway tunnels. The workflow includes automated data acquisition and systematisation, algorithmic generation of structured, data-rich BIM models, and a centralised digital twin platform for lifecycle monitoring.Findings Implementation enabled real-time data manipulation, predictive maintenance and improved decision-making. Key outputs include standardised data exchange protocols, product data templates, BIM object classes for tunnel components and digital twinning algorithms. Benefits observed include improved collaboration, automatic anomaly detection and enhanced visualisation of tunnel condition.Research limitations/implications Further research is needed to improve the algorithm's performance, minimise manual interventions and validate the workflow's scalability across diverse asset typologies and operational contexts.Originality/value The research presents a framework for integrating digital twins into infrastructure asset management, demonstrating the operational value of computational methods in transforming conventional maintenance and lifecycle management workflows and supporting data-driven decision-making.