Use of big data analytical techniques and machine learning for monitoring and maintaining civil engineering infrastructure

Over their design lives, pieces of civil engineering infrastructure are expected to experience some fatigue. However, the effects of climate change and increased public usage of infrastructure (e.g. roads, railways) are exerting greater stresses on structural elements (e.g. foundations, retaining walls, embankments/cuttings), thereby causing them to degrade at a faster rate and potentially lead to premature failures. Such failures cause major disruption to the infrastructure network and are costly to repair, from financial and energy (environmental) standpoints. Hence, there is a need to install and monitor instrumentation within infrastructure assets during and post-construction and the use of big data analytics/machine learning techniques, with a view to making: 1) predictions of any potential future failures and 2) data-informed decisions for undertaking asset maintenance/repair work (rather than using a time-based approach for inspecting assets). This project focusses on the use of geotechnical monitoring data collected from a real-life highway case study.