Nenad Gucunski

The deterioration of concrete bridge decks is a long-term complex phenomenon that is difficult to examine using traditional inspection methods. Variables such as environmental exposure and traffic volume affect the condition of the deck throughout the decades of its design lifecycle. Hence, it is paramount to develop a framework through which a deck’s condition can be monitored from construction to rehabilitation. Accelerated structural testing combined with the use of nondestructive evaluation can be developed as a knowledge-creation method to inform the initiation, progression, and propagation of deterioration in bridge decks in a compressed time frame. In this paper, the results of such a framework are presented to introduce the concept of large-scale accelerated structural testing and deck condition monitoring. The BEAST (Bridge Evaluation and Accelerated Structural Testing) facility at Rutgers University was constructed for this purpose and was deployed in the current study. Several nondestructive evaluation techniques were used to monitor the condition of the deck throughout the experimental program, including GPR, IE, USW, HCP, and ER. Condition maps and indices tracking the progression of deterioration in a bare concrete deck are presented. In addition, results from deck segmentation on fundamental statistics from different NDE measurements are shown in the current study. Over 843 days, the specimen was subjected to 2.05 million live-load passes, 98 freeze–thaw cycles, and 115 brine applications. The HCP condition index declined to approximately 60, the average dynamic modulus decreased from a peak of approximately 36 GPa to approximately 30 GPa, and deterioration became concentrated in the reduced-cover region.

When a structural health assessment is conducted on typical bridges, bridge decks are commonly found to be the fastest deteriorating component of the bridge. This is the case since bridge decks are the most exposed component of the bridge to various environmental elements and traffic loads. Traveling vehicles on the bridge, namely trucks, impart cyclic stresses on the bridge deck depending on their location and bridge occupancy. The continuous reversal of stress between tensile and compressive stress enveloped by a stress magnitude leads to fatigue, and ultimately the deterioration of the bridge deck, which can be exacerbated by environmental factors. Despite some bridges having similar bridge decks, some would deteriorate faster than others, even if they have a similar deck and exposure conditions. In this case, the supporting superstructure rigidity becomes key in understanding why different deterioration/damage levels are exhibited. This research assesses the effects of having varied superstructural rigidity on the deck response due to operational load levels. It provides insights into the stresses of bridge decks due to moving vehicles and vibration magnitudes.

The performance evaluation of a concrete bridge deck with ultra-high-performance concrete (UHPC) and Latex-modified concrete (LMC) overlay was evaluated in the BEAST (Bridge Evaluation and Accelerated Structural Testing) facility at Rutgers University. A bridge structure 15 m long and 8.5 m wide was subjected to accelerated live and environmental loadings. The accelerated live loading is achieved by continuously passing a dual-axle carriage, applying a load of 230 kN over the deck. More than 10,000 carriage passes are made per day. In addition, the environmental loads are simulated through rapid freeze-thaw cycling and spreading of salt brine. The deck's condition is evaluated after approximately 200,000 live-loading cycles by five nondestructive evaluation (NDE) technologies: electrical resistivity, half-cell potential, impact echo, ultrasonic surface waves, and ground penetrating radar. After more than four million live-loading cycles, the UHPC- and LMC-overlaid deck showed superior performance over the bare concrete deck tested with only two million live-loading cycles concerning corrosion, delamination formation, and concrete quality degradation. The results of the study also demonstrate the capability of the BEAST facility, combined with NDE, to induce, capture, and quantify deterioration and defect formation processes in bridge structures in an accelerated manner. The paper provides an overview of the BEAST facility, the bridge deck condition assessment and long-term monitoring by NDE technologies, and a comparison of the performance results of the bare and overlaid bridge decks and corresponding anticipated performance on actual bridges.

Monitoring the condition of concrete bridge decks is essential for maintaining performance, extending service life, and supporting sustainable infrastructure. This study presents a novel machine learning (ML) approach for assessing concrete deck deterioration by analyzing spatial statistical features of electrical resistivity (ER) data. ER data were collected over 850 days through 14 measurement cycles in a full-scale accelerated bridge testing program at Rutgers University, designed to simulate realistic field conditions. Concrete deterioration was evaluated using dynamic modulus values obtained from ultrasonic surface wave measurements. Principal spatial features—average, coefficient of variation, and entropy—extracted from ER data, corrected for non-destructive factors (temperature, age, and salinity), were employed to train various ML models. The support vector machine (SVM) achieved an F1 score of 0.82, surpassing traditional threshold methods by 60%. The results demonstrate the effectiveness of this ML approach in tracking damage progression and enabling early anomaly detection. These findings are significant for network-level assessments of large bridge decks and provide a foundation for more robust deterioration assessment frameworks.

Several factors impact the durability of concrete bridge decks, including traffic loads, fatigue, temperature changes, environmental stress, and maintenance activities. Detecting problems such as corrosion, delamination, or concrete degradation early on can lower maintenance costs. Nondestructive evaluation (NDE) techniques can detect these issues at early stages. Each NDE method, meanwhile, has limitations that reduce the accuracy of the assessment. In this study, multiple NDE technologies were combined with machine learning algorithms to improve the interpretation of half-cell potential (HCP) and electrical resistivity (ER) measurements. A parametric study was performed to analyze the influence of five parameters on HCP and ER measurements, such as the degree of saturation, corrosion length, delamination depth, concrete cover, and moisture condition of delamination. The results were obtained through finite element simulations and used to build two machine learning algorithms, a classification algorithm and a regression algorithm, based on Random Forest methodology. The algorithms were tested using data collected from a bridge deck in the BEAST^® facility. Both machine learning algorithms were effective in improving the interpretation of the ER and HCP measurements using data from multiple NDE technologies.

Several factors impact the durability of concrete bridge decks, including traffic loads, fatigue, temperature changes, environmental stress, and maintenance activities. Detecting problems such as corrosion, delamination, or concrete degradation early on can lower maintenance costs. Nondestructive evaluation (NDE) techniques can detect these issues at early stages. Each NDE method, meanwhile, has limitations that reduce the accuracy of the assessment. In this study, multiple NDE technologies were combined with machine learning algorithms to improve the interpretation of half-cell potential (HCP) and electrical resistivity (ER) measurements. A parametric study was performed to analyze the influence of five parameters on HCP and ER measurements, such as degree of saturation, corrosion length, delamination depth, concrete cover, and moisture condition of delamination. The results were obtained through finite element simulations and used to build two machine learning algorithms, a classification algorithm and a regression algorithm, based on the Random Forest methodology. The algorithms were tested using data collected from a bridge deck in the BEAST® facility. Both machine learning algorithms were effective in improving the interpretation of ER and HCP measurements using data from multiple NDE technologies.