Developing a scientific decision optimisation framework is essential for improving resource efficiency in bridge maintenance. This study proposes a multi-level group maintenance optimisation framework for bridge systems based on a genetic algorithm with agglomerative mutation (GA-A). The framework integrates component-level deterioration simulation using a continuous-time Markov chain, system-level cost savings considering dual economic-structural correlations, and optimisation of dynamic group scheduling. By incorporating an agglomerative mutation operator, the GA-A algorithm enables adaptive regrouping, effectively addressing the non-deterministic polynomial (NP)-hard problem of dynamic maintenance scheduling. A case study on a bridge system demonstrates that the GA-A-based strategy significantly outperforms independent maintenance approaches, achieving higher cost savings and lower system downtime, particularly under strong structural correlations. When the structural correlation coefficient increases from 0 to 1.0, the optimal group maintenance strategy for the bridge system achieves cost reductions ranging from 6.68% to 19.35% compared with the independent maintenance strategy. The proposed framework effectively addresses current gaps in cost-savings assessment, resource allocation and dynamic scheduling for bridge system maintenance.

In the seismic reliability assessment of regional bridge networks (RBNs), traditional connectivity reliability (CR) mainly measures the ability of bridge networks to maintain physical connectivity under seismic effects. Meanwhile, travel time (Tt) reflects the change in access efficiency after an earthquake. However, reliance on a single indicator can lead to distorted assessment results. This study proposes a method to assess the seismic reliability of regional bridge networks based on the composite index (TtCR). The proposed index integrates the structural and traffic characteristics of the RBN to provide a comprehensive assessment of its seismic reliability. Finally, an actual bridge network in South Dakota is used as a case study. The results show that under minor earthquakes (PGA = 0.2g), TtCR exceeds Tt by 33%, indicating its ability to mitigate over-pessimistic estimations caused by solely considering Tt. For moderate earthquakes (PGA = 0.6g), TtCR exhibits a 128% increase over Tt but remains 37% lower than CR, striking a balance between these two extreme metrics. In strong earthquakes (PGA = 1.0g), TtCR is approximately five times larger than Tt, ensuring reliable assessments even under extreme conditions by avoiding undue pessimism from low Tt values.

Maritime ports are key components of global logistics networks, with steel quay walls providing berthing capacity and operational continuity. Their long-term structural performance is governed by corrosion driven by interactions between salinity, hydrodynamics, microbiological activity, and climatic conditions. Given that across Europe, many twentieth-century structures have exceeded their design life, reassessment of safety and residual capacity is essential. Conventional assessments typically use deterministic, uniform corrosion profiles based on simplified environmental classifications. In practice, however, field data show that corrosion is spatially variable, has short correlation lengths, and involves co-existing uniform and localised mechanisms. The scarcity of long-term, spatially detailed measurements has limited of site-specific deterioration models to be validated and included in design codes. This study analyses corrosion in steel quay walls at the Port of Rotterdam using ultrasonic thickness measurements and laboratory surface-morphology data. The database quantifies mean wall-thickness loss and spatial variability, enabling systematic comparison with design prescriptions. To interpret the observed variability, the study develops a stochastic corrosion representation based on random-fields, allowing explicit incorporation of spatial heterogeneity into structural assessments. The outcomes highlight the limitations of uniform corrosion assumptions and provide a basis for improved reliability evaluations and lifecycle-management strategies for ageing port infrastructure.

With the increasing demand for highway transportation, the structural integrity of ageing bridges has become a critical concern. Transverse steel connections (TSCs) offer distinct advantages in economic efficiency and rapid construction. This study introduced a novel H-shaped TSC and proposed arrangement improvements to reduce costs. A field loading test on a four-span continuous multi-box girder bridge showed that the H-shaped TSC increased vertical bending frequency by 16.2% and reduced peak girder deflection by 5.7%. A finite element (FE) model, built in ABAQUS and validated by test results, was used for parametric analyses. Results indicated that structural form and vertical height had negligible effects, while steel consumption was the dominant factor; efficiency dropped significantly when the cross-sectional area fell below 12.5%. A systematic TSC design methodology was developed and validated, providing a practical framework for optimising arrangement and configuration. These findings offer recommendations for reinforcing multi-box girder bridges.

As critical parts of transportation networks, bridges are increasingly vulnerable to vehicle explosions caused by military conflicts, terrorist attacks, or accidental detonations. Understanding such effects is essential for enhancing bridge safety and minimising economic loss. However, existing studies often use scenario-based analyses with deterministic blast hazards, neglecting uncertainties in hazard occurrence due to limited data. Bridge restoration following vehicle explosions also faces challenges. This study addresses these gaps by introducing a probabilistic approach to evaluate the life-cycle resilience and economic loss of bridges exposed to under-deck vehicle explosions. It combines a fuzzy logic-based method to estimate the stochastic occurrence of explosions, Monte Carlo Simulation to analytically quantify blast-induced failure probabilities, and the Poisson Renewal Process to evaluate long-term resilience and loss. The proposed approach is illustrated through a case study of a reinforced concrete bridge, analysing explosions involving passenger vehicles, freight trucks, and fuel tankers. Results show that fuel tanker explosions lead to the most severe structural damage and lowest resilience, with significant deterioration over the life-cycle. Freight truck explosions, while less damaging structurally, result in the highest long-term economic risk. Higher discount rates slow the accumulation of economic loss, whereas larger baseline costs substantially increase financial burdens.

This study investigates the use of External Adaptive Tensioning (EAT) systems as a retrofit strategy to reduce the structural response and fatigue damage in steel bridges. A high-fidelity three-dimensional model, including detailed welded-joint sub-models, is combined with the Active System Utilisation (ASU) metric to account for actuator reliability and fallback configurations. Results show that EAT reduces the stress response by up to 66% relative to the unretrofitted condition and achieves near-zero fatigue damage at critical welded details under representative traffic loading. Even with partial downtime (ASU < 1), EAT provides substantial gains in remaining service life compared with passive external post-tensioning. These findings demonstrate the structural benefits of retrofitting with active components and establish a computational framework for evaluating fatigue and reliability effects in actively controlled bridge systems.

Previous seismic events have emphasised the critical role of beam-column joints in the performance of moment-resisting frames. This study investigates retrofitting techniques for non-ductile reinforced concrete beam-column joints under cyclic loading, evaluating structural performance, environmental sustainability, and economic feasibility. Finite element models were developed and validated using ABAQUS simulations. A parametric study targeting a 37 kN load capacity identified three effective methods: CFRP grids with engineered cementitious composites (ECC), CFRP sheets with steel anchor bolts, and steel jacketing. Among these, CFRP grids with ECC provided the highest structural performance by mitigating failure modes such as cracking. CFRP sheets with steel anchor bolts and steel jacketing also improved performance but exhibited limitations like stress concentration and plastic deformation. A life cycle assessment (LCA) conducted with SimaPro® showed that CFRP sheets with steel anchors had the lowest global warming potential and human health impact. Despite its higher environmental footprint, steel jacketing proved most cost-effective in the life cycle cost (LCC) analysis. A multi-criteria decision-making (MCDM) approach integrating LCA and LCC through the weighted sum method (WSM) ultimately recommended steel jacketing as the optimal retrofitting solution due to its balanced performance, durability, and economic advantages.

Continuous bridge monitoring generates long-duration, noisy, and multi-channel acceleration streams in which abnormal events are sparse yet safety-critical. This work targets reliable event classification under operational conditions while preserving the time-series nature of the data. A hybrid one-dimensional convolutional model with channel-wise recalibration and attention analyzes fixed-length windows from a cable-stayed bridge monitoring system, combined with peak picking, batching, and lightweight inference that suit day-scale records. Evaluation on a held-out test set shows balanced class performance: precision across classes ranges from 90.5% to 98.4%, and recall across classes ranges from 91.4% to 99.6%, yielding precision of 94.6% and recall of 94.9%. At the sensor level, classification accuracy ranges from 93.6% to 99.7%, indicating consistent behaviour across heterogeneous channels with different noise and placement conditions. The dataset is split into training, validation, and test subsets, and the test set remains untouched during model selection. The results suggest robust detection across classes and sensors, stable operation on continuous 24-h records, and practical readiness for field deployment in bridge monitoring.

The accelerated ageing of reinforced concrete (RC) due to environmental exposure creates significant economic risks, demanding timely, objective condition assessment. The paper presents a methodology that screens and classifies degradation indicators (DI), like carbonation depth, chloride content, corrosion potential, and electrical resistivity, into four damage classes, then uses transition functions to map classes to continuous values. These are combined through the Analytic Hierarchy Process (AHP) to compute a Corrosion Damage Index (CDI), which enters a Fuzzy‑Logic (FL) decision system for selecting repair principles. The fuzzy inference comprises fuzzification, expert rule‑based inference, and defuzzification to evaluate the optimal rehabilitation options. Practicality was demonstrated on a RC retaining wall in Vienna using real inspection data. The wall was discretised into a grid; DI were measured at grid nodes to represent the survey input. Measurements were converted to CDI via AHP and fed to the FL system to produce a spatial map of potential interventions at each grid point, providing location‑specific guidance for maintenance. The approach scales from component to network level, supporting service‑life forecasting and budget optimisation through aggregated CDI trends and intervention performance. Integrating expert knowledge with quantitative analysis, the system offers a robust, sustainable, and data‑driven framework for planning and prioritising rehabilitation.