Incorporating Digital Twins and Artificial Intelligence for Next-Generation SHM Software

The management of aging built infrastructure represents one of the most pressing challenges facing most developed countries. The circumstance that a sizable portion of the world’s infrastructure, mostly made of reinforced concrete, is approaching its design life-expectancy of 50–100 years, poses a serious risk of cascade failures with severe socio-economic consequences. This has led to the growing implementation of new maintenance regulations, in which Structural Health Monitoring (SHM) plays predominant role as an effective means of preventive maintenance. Nonetheless, the extensive implementation of SHM systems still encounters serious difficulties, both from hardware and software standpoints. Clear evidence of the latter is the current lack of widely accepted software programs capable of managing dense sensor networks on a regional scale. Indeed, most innovations in the realm of SHM remain confined to research environments without a clear transition to engineering routine practice. In this context, this work presents the latest advances in the development of a comprehensive SHM software platform called MOVA/MOSS. Originally designed as a software framework for conducting long-term integrated SHM of structures (encompassing both dynamic and static monitoring), the most recent upgrades involve the incorporation of digital twins and artificial intelligence for real-time system and damage identification. The potential of the developed software platform is showcased through several real-world structures, with particular focus on the integration possibilities of non-supervised (data-driven) and supervised (model-driven) damage identification.