This paper presents an approach for evaluating concrete cracking growth and structural resistance deterioration due to reinforcement corrosion and for predicting future structural performance during the service life of the corroded concrete beams. The evolution of corrosion-induced concrete cracking is analytically predicted and then adopted for estimating the residual bond strength of the corroded steel rebar. The flexural capacity deterioration of a corroded concrete beam is accurately determined with consideration of rebar cross-section loss, yielding strength reduction and bond strength degradation caused by reinforcement corrosion. The concrete cracking growth and flexural capacity deterioration rates are chosen as random variables for modeling structural performance deterioration over time during the life cycle. The gamma process is then used for time-dependent stochastic deterioration modeling and the remaining life estimate of the corroded concrete beams. The results for the worked example show that the proposed approach can provide reliable predictions for concrete crack development and structural strength deterioration, and also give an effective method for lifetime performance assessment of the corroded concrete structures.

Structural Health Monitoring (SHM) has emerged as a viable alternative to traditional visual and non-destructive assessment methods for civil infrastructure. The integration of sensor systems, the Internet of Things (IoT), and advanced data processing has further digitised SHM, leading to the development of Digital Twin (DT) technology, enabling dynamic, real-time simulations for proactive risk prediction and asset management. However, many existing DT-based SHM systems are costly, complex, and resource-intensive, limiting their practicality for small-scale implementations. This study investigates the feasibility of a practical and cost-effective SHM framework enhanced by DT technology for civil infrastructure. The hardware system explored two low-cost displacement sensors: a potentiometer contact sensor and an infrared non-contact sensor. During the static load testing, the potentiometer demonstrated high accuracy and stability, whilst the infrared sensor, despite higher noise, was effective for submillimetre measurements. These sensors were integrated with an IoT-enabled Arduino Nano 33 microcontroller for remote access via the cloud platform. The software system, “ReConTwin”, developed using open-source resources, provides near real-time updates, analysis, and damage diagnosis through an automated post-processing system. The calibrated DT replicated the force–displacement response, accurately estimated the applied load, and closely predicted mid-span strain and crack formations of a Reinforced Concrete (RC) beam specimen subjected to short-term static three-point bending loads in a controlled laboratory setting. The user-friendly design and compatibility with standard commercial computers enhance the accessibility and feasibility of the proposed DT-SHM framework, making it a promising scalable solution for widespread adoption in real-world civil infrastructure applications.