Infrastructures

14 pages, 1515 KB

Open AccessArticle

Zero-Shot Bridge Health Monitoring Using Cepstral Features and Streaming LSTM Networks

by Azin Mehrjoo, Kyle L. Hom, Homayoon Beigi and Raimondo Betti

Infrastructures 2025, 10(11), 292; https://doi.org/10.3390/infrastructures10110292 - 3 Nov 2025

This paper presents a real-time, output-only structural health monitoring framework that integrates damage-sensitive cepstral features with a streaming Long Short-Term Memory (LSTM) network for automated damage detection. Acceleration time histories are segmented into overlapping windows, converted into cepstral coefficients, and processed sequentially by [...] Read more.

This paper presents a real-time, output-only structural health monitoring framework that integrates damage-sensitive cepstral features with a streaming Long Short-Term Memory (LSTM) network for automated damage detection. Acceleration time histories are segmented into overlapping windows, converted into cepstral coefficients, and processed sequentially by a stacked LSTM architecture with state carry-over. This design preserves temporal dependencies while enabling low-latency inference suitable for continuous monitoring. The framework was evaluated under a strict zero-shot setting on the full-scale Z24 Bridge benchmark, in which no training or calibration data from the bridge were used. Our results show that the proposed approach can reliably discriminate staged damage states and track their progression using only vibration measurements. By combining a well-established spectral feature representation with a streaming sequence model, the study demonstrates a practical pathway toward deployable, data-driven monitoring systems capable of operating without retraining on each individual asset. Full article

(This article belongs to the Special Issue Advances in Structural Health Monitoring and Industry 5.0 Innovations for Bridge Management and Conservation)

► Show Figures

Figure 1

18 pages, 4298 KB

Open AccessArticle

Life-Cycle-Assessment-Based Quantification and Low-Carbon Optimization of Carbon Emissions in Expressway Construction

by Zhen Liu

Infrastructures 2025, 10(11), 291; https://doi.org/10.3390/infrastructures10110291 - 2 Nov 2025

Abstract

To quantitatively assess the carbon emission characteristics of expressway construction and to identify its key influencing factors, this study establishes a comprehensive carbon emission accounting framework that covers the material production, transportation, and construction stages based on the life cycle assessment (LCA) approach. [...] Read more.

To quantitatively assess the carbon emission characteristics of expressway construction and to identify its key influencing factors, this study establishes a comprehensive carbon emission accounting framework that covers the material production, transportation, and construction stages based on the life cycle assessment (LCA) approach. Typical expressway projects are selected as case studies to perform stage-based emission quantification and multivariable response analysis. The results indicate that the total carbon emissions per kilometer during the construction phase are approximately 1.80 × 10³ kg CO₂-eq/km, with material production being the dominant contributor, accounting for about 60–70%, followed by transportation and construction activities. The analysis of structural layers shows that variations in the thickness of the asphalt surface and cement-stabilized base layers, which are the main sources of emissions, are strongly and positively correlated with total emissions, making them the principal control factors. Transportation distance and equipment efficiency are identified as moderately sensitive parameters, each contributing approximately 3–5% to emission variation. Further multivariable response analysis demonstrates nonlinear coupling effects between structural parameters and transportation factors. The combined increase in layer thickness and transport distance significantly amplifies total emissions, while the marginal impact of long-distance transport gradually decreases. Based on these findings, this study proposes a low-carbon construction strategy that focuses on structural optimization, local material sourcing, energy-efficient construction practices, and the use of clean energy. The outcomes of this research provide a theoretical foundation and quantitative reference for carbon emission prediction, structural design optimization, and green construction decision making during the expressway construction phase. Full article

(This article belongs to the Special Issue Sustainable Road Design and Traffic Management)

► Show Figures

Figure 1

21 pages, 1267 KB

Open AccessReview

More Effective Front-End Decision-Making for Pipe Renewal Projects

by Bjørn Solnes Skaar, Tor Kristian Stevik, Agnar Johansen and Asmamaw Tadege Shiferaw

Infrastructures 2025, 10(11), 290; https://doi.org/10.3390/infrastructures10110290 - 31 Oct 2025

Viewed by 114

Abstract

Access to clean, hygienic, and sufficient potable water is a concern in many countries. To ensure this, asset management, planning, and structured pipe renewal are crucial in providing an adequate level of service. However, there is a significant backlog in municipal pipe renewal, [...] Read more.

Access to clean, hygienic, and sufficient potable water is a concern in many countries. To ensure this, asset management, planning, and structured pipe renewal are crucial in providing an adequate level of service. However, there is a significant backlog in municipal pipe renewal, which needs to be addressed to raise the standard of potable water supply to an acceptable level in countries across most continents. Therefore, the objective of this research was to improve decision-making to reduce this backlog. Competent personnel are a scarce resource and not easily replaced. Standardized decision-making is considered an efficient approach to addressing the shortage of skilled personnel in pipe renewal. However, its effectiveness depends on its adaptability to the varying complexity and scale of such projects during implementation. This research is based on a literature review that explores decision theories, project definitions, and project models, and compares the typical characteristics of pipe renewal projects with those of other infrastructure projects. The research highlights that structured and standardized decision-making processes are essential to ensure appropriate asset management of the pipe network and sufficient pipe renewal. The main outcome of this research is a tailored project model that supports better front-end decision-making in pipe renewal projects through improved information flow. Full article

► Show Figures

Figure 1

22 pages, 2777 KB

Open AccessArticle

Efficient Dual-Domain Collaborative Enhancement Method for Low-Light Images in Architectural Scenes

by Jing Pu, Wei Shi, Dong Luo, Guofei Zhang, Zhixun Xie, Wanying Liu and Bincan Liu

Infrastructures 2025, 10(11), 289; https://doi.org/10.3390/infrastructures10110289 - 31 Oct 2025

Viewed by 59

Abstract

Low-light image enhancement in architectural scenes presents a considerable challenge for computer vision applications in construction engineering. Images captured in architectural settings during nighttime or under inadequate illumination often suffer from noise interference, low-light blurring, and obscured structural features. Although low-light image enhancement [...] Read more.

Low-light image enhancement in architectural scenes presents a considerable challenge for computer vision applications in construction engineering. Images captured in architectural settings during nighttime or under inadequate illumination often suffer from noise interference, low-light blurring, and obscured structural features. Although low-light image enhancement and deblurring are intrinsically linked when emphasizing architectural defects, conventional image restoration methods generally treat these tasks as separate entities. This paper introduces an efficient and robust Frequency-Space Recovery Network (FSRNet), specifically designed for low-light image enhancement in architectural contexts, tailored to the unique characteristics of such scenes. The encoder utilizes a Feature Refinement Feedforward Network (FRFN) to achieve precise enhancement of defect features while dynamically mitigating background redundancy. Coupled with a Frequency Response Module, it modifies the amplitude spectrum to amplify high-frequency components of defects and ensure balanced global illumination. The decoder utilizes InceptionDWConv2d modules to capture multi-directional and multi-scale features of cracks. When combined with a gating mechanism, it dynamically suppresses noise, restores the spatial continuity of defects, and eliminates blurring. This method also reduces computational costs in terms of parameters and MAC operations. To assess the effectiveness of the proposed approach in architectural contexts, this paper conducts a comprehensive study using low-light defect images from indoor concrete walls as a representative case. Experimental results indicate that FSRNet not only achieves state-of-the-art PSNR performance of 27.58 dB but also enhances the mAP of the downstream YOLOv8 detection model by 7.1%, while utilizing only 3.75 M parameters and 8.8 GMACs. These findings fully validate the superiority and practicality of the proposed method for low-light image enhancement tasks in architectural settings. Full article

36 pages, 3299 KB

Open AccessArticle

Mechanistic-Empirical Analysis of LDPE-SBS-Modified Asphalt Concrete Mix with RAP Subjected to Various Traffic and Climatic Loading Conditions

by Muhammad Haris, Asad Naseem, Sarfraz Ahmed, Muhammad Kashif and Ahsan Naseem

Infrastructures 2025, 10(11), 288; https://doi.org/10.3390/infrastructures10110288 - 30 Oct 2025

Viewed by 157

Abstract

The current global economic challenges and resource scarcity necessitate the development of cost-effective and sustainable pavement solutions. This study investigates the performance of asphalt mixtures modified with Low-Density Polyethylene (LDPE) and Styrene–Butadiene–Styrene (SBS) as binder modifiers, and Hydrated Lime (Ca(OH)₂) and [...] Read more.

The current global economic challenges and resource scarcity necessitate the development of cost-effective and sustainable pavement solutions. This study investigates the performance of asphalt mixtures modified with Low-Density Polyethylene (LDPE) and Styrene–Butadiene–Styrene (SBS) as binder modifiers, and Hydrated Lime (Ca(OH)₂) and Reclaimed Asphalt Pavement (RAP) as aggregate replacements. The research aims to optimize the combination of these materials for enhancing the durability, sustainability, and mechanical properties of asphalt mixtures under various climatic and traffic conditions. Asphalt mixtures were modified with 5% LDPE and 2–6% SBS (by bitumen weight), with 2% Hydrated Lime and 15% RAP added to the mix. The performance of these mixtures was evaluated using the Simple Performance Tester (SPT), focusing on rutting, cracking, and fatigue resistance at varying temperatures and loading frequencies. The NCHRP 09-29 Master Solver was employed to generate master curves for input into the AASHTOWare Mechanistic-Empirical Pavement Design Guide (MEPDG), allowing for an in-depth analysis of the modified mixes under different traffic and climatic conditions. Results indicated that the mix containing 5% LDPE, 2% SBS, 2% Hydrated Lime, and 15% RAP achieved the best performance, reducing rutting, fatigue cracking, and the International Roughness Index (IRI), and improving overall pavement durability. The combination of these modifiers showed enhanced moisture resistance, high-temperature rutting resistance, and improved dynamic modulus. Notably, the study revealed that in warm climates, thicker pavements with this optimal mix exhibited reduced permanent deformation and better fatigue resistance, while in cold climates, the inclusion of 2% SBS further improved the mix’s low-temperature performance. The findings suggest that the incorporation of LDPE, SBS, Hydrated Lime, and RAP offers a sustainable and cost-effective solution for improving the mechanical properties and lifespan of asphalt pavements. Full article

(This article belongs to the Special Issue Safer Roads Ahead: Exploring the Latest Innovations and Advancements in Road Design and Safety Technology)

► Show Figures

Figure 1

16 pages, 2880 KB

Open AccessArticle

Comparative Assessment of Vertical Precision of Unmanned Aerial Vehicle-Based Geodetic Survey for Road Construction: A Multi-Platform and Multi-Software Approach

by Brankica Malić, Vladimir Moser, Damir Rajle, Saša Kulić and Ivana Barišić

Infrastructures 2025, 10(11), 287; https://doi.org/10.3390/infrastructures10110287 - 30 Oct 2025

Viewed by 137

Abstract

Accurate geodetic surveys are essential for road design, with altimetric accuracy being particularly critical. UAV photogrammetry offers faster and safer data acquisition than conventional methods, but its applicability depends on whether it can meet engineering accuracy standards. This study investigates the altimetric accuracy [...] Read more.

Accurate geodetic surveys are essential for road design, with altimetric accuracy being particularly critical. UAV photogrammetry offers faster and safer data acquisition than conventional methods, but its applicability depends on whether it can meet engineering accuracy standards. This study investigates the altimetric accuracy of UAV photogrammetry through a comparative assessment of surveys conducted on the same urban roundabout in Osijek, Croatia, in 2016 and 2024. By conducting the surveys eight years apart at the same location, the study allows for an assessment of how technological and methodological developments affect survey outcomes. The research evaluates different UAVs and multiple SfM software packages in a comparative framework, highlighting how UAV–software combinations affect results, rather than attributing accuracy solely to hardware or processing. The results of the conducted research indicate a significant increase in the accuracy of the UAV photogrammetric survey method. Through a proper combination of UAVs and SfM processing software, it is possible to achieve an accuracy within 2 cm and an RMSE of 1.2 cm, which is in line with the accuracy of a standard survey method like GNSS CROPOS. The results underline that UAV photogrammetry, when properly planned and executed, can now deliver altimetric accuracy sufficient for most road construction tasks, providing a reliable and cost-effective alternative to conventional geodetic surveys. Full article

(This article belongs to the Special Issue Safer Roads Ahead: Exploring the Latest Innovations and Advancements in Road Design and Safety Technology)

► Show Figures

Figure 1

24 pages, 7976 KB

Open AccessArticle

Experimental and Numerical Model Analysis of Pipe–Soil Interaction Under Typical Geohazard Conditions

by Ning Shi, Tianwei Kong, Xiaoben Liu and Hong Zhang

Infrastructures 2025, 10(11), 286; https://doi.org/10.3390/infrastructures10110286 - 29 Oct 2025

Viewed by 111

Abstract

This paper systematically investigates the interaction between pipes and soil under geo-logical disaster conditions by combining small-scale physical experiments with mul-ti-method numerical simulations. Three analytical models—namely the Smoothed Particle Hydrodynamics-Finite Element Method (SPH-FEM) model, the traditional FEM model, and the soil spring-based Pipe–Soil [...] Read more.

This paper systematically investigates the interaction between pipes and soil under geo-logical disaster conditions by combining small-scale physical experiments with mul-ti-method numerical simulations. Three analytical models—namely the Smoothed Particle Hydrodynamics-Finite Element Method (SPH-FEM) model, the traditional FEM model, and the soil spring-based Pipe–Soil Interaction (PSI) model—are employed to comparatively analyze their applicability across different geohazard scenarios. The study found that the PSI model overpredicted pipeline strain responses, indicating that traditional soil spring analytical models require modification. The traditional FEM model provided the most accurate predictions under small-displacement conditions, while the SPH-FEM model yielded more reliable results for large-displacement scenarios. The novelty of this study lies in its systematic exploration of the applicability of these three methodologies, providing scientifically grounded simulation tools for numerical modeling in engineering practice. Full article

► Show Figures

Figure 1

22 pages, 6011 KB

Open AccessArticle

Effect of Stochastic Guideway Irregularity on Dynamic Performance of Maglev Train

by Tian Qin, Deqiu Kong, Yang Song, Like Pan and Cheng Zhang

Infrastructures 2025, 10(11), 285; https://doi.org/10.3390/infrastructures10110285 - 27 Oct 2025

Viewed by 131

Abstract

Maglev trains represent an advanced form of modern rail transportation. The guideway irregularity presents a common disturbance to the safe and reliable operation of the maglev train. Variations in the air gap between the train and the guideway, induced by the guideway irregularities, [...] Read more.

Maglev trains represent an advanced form of modern rail transportation. The guideway irregularity presents a common disturbance to the safe and reliable operation of the maglev train. Variations in the air gap between the train and the guideway, induced by the guideway irregularities, exert a significant influence on the train’s dynamic performance, thereby impacting both ride comfort and operational safety. Although previous studies have acknowledged the importance of guideway irregularity, the stochastic effects on the car body vibration across different speeds have not been quantitatively assessed. To fill in this gap, this paper presents a 10-degree-of-freedom maglev train model based on multibody dynamics. The guideway is modelled via the finite element method using Euler–Bernoulli beam theory, and a linearized electromagnetic force equation is employed to couple the guideway and the train dynamics. Furthermore, the measurement data of guideway irregularity from the Shanghai Maglev commercial line are incorporated to evaluate their stochastic effect. Analysis results under varying speeds and irregularity wavelengths identify a resonance speed of 127.34 km/h, attributed to the interplay between guideway periodicity and the train’s natural frequency. When the ratio of the train speed versus irregularity wavelength satisfies the train’s natural frequency, a significant resonance can be observed, leading to an increase in train vibration. Based on the Monte Carlo method, stochastic analysis is conducted using 150 simulations per speed in 200–600 km/h. The maximum vertical acceleration remains relatively stable at 200–400 km/h but increases significantly at higher speeds. When the irregularity is present, greater dispersion is observed with increasing speed, with the standard deviation at 600 km/h reaching 2.7 times that at 200 km/h. Across all tested cases, acceleration values are consistently higher than those without irregularities within the corresponding confidence intervals. Full article

(This article belongs to the Special Issue The Resilience of Railway Networks: Enhancing Safety and Robustness)

► Show Figures

Figure 1

19 pages, 2844 KB

Open AccessArticle

Statistical Analysis of the Tensile Strength of Cold Recycled Cement-Treated Materials and Its Influence on Pavement Design

by William Fedrigo, Thaís Radünz Kleinert, Gabriel Grassioli Schreinert, Lélio Antônio Teixeira Brito and Washington Peres Núñez

Infrastructures 2025, 10(11), 284; https://doi.org/10.3390/infrastructures10110284 - 24 Oct 2025

Viewed by 276

Abstract

The tensile behavior of cold recycled cement-treated mixtures (CRCTMs), typically produced through full-depth reclamation (FDR), is critical for pavement design. Since no universal design method exists, different tests are applied, leading to varying results. In this context, this study aimed (a) to statistically [...] Read more.

The tensile behavior of cold recycled cement-treated mixtures (CRCTMs), typically produced through full-depth reclamation (FDR), is critical for pavement design. Since no universal design method exists, different tests are applied, leading to varying results. In this context, this study aimed (a) to statistically analyze the flexural tensile strength (FTS) and indirect tensile strength (ITS) of CRCTMs incorporating reclaimed asphalt pavement (RAP) and lateritic soil (LS); (b) to evaluate how using FTS or ITS influences the design of CRCTM layers. FTS and ITS tests were conducted with different cement (1–7%) and RAP (7–93%) contents at multiple curing times (3–28 days), and results were used for statistical and mechanistic analyses. Results showed that cement and RAP contents significantly increased FTS and ITS. RAP exhibited the strongest influence on ITS. This indicates that CRCTMs with similar materials benefit from higher RAP contents. Mechanistic analysis revealed that lower RAP contents require thicker pavement structures, suggesting that increasing RAP can reduce costs and environmental impacts. FTS was about 65% higher than ITS, but using ITS in design led to structures 1.7–3.3 times thicker for the same service life. These findings highlight the need for proper CRCTM characterization, with flexural tests recommended for more reliable and cost-effective pavement design. Full article

(This article belongs to the Special Issue Cold and Warm Techniques for Sustainable Pavement Construction and Maintenance)

► Show Figures

Figure 1

25 pages, 16408 KB

Open AccessArticle

Understanding Pavement Texture Evolution and Its Impact on Skid Resistance Through Machine Learning

by Yiwen Zou, Guanliang Chen, Guangwei Yang and Xu Chen

Infrastructures 2025, 10(11), 283; https://doi.org/10.3390/infrastructures10110283 - 24 Oct 2025

Viewed by 245

Abstract

The texture of asphalt pavement wears down over time due to traffic polishing, which leads to polished pavement surfaces with lower skid resistance. Three-dimensional (3D) texture parameters can be used to describe the evolution of pavement texture and establish predictive models for skid [...] Read more.

The texture of asphalt pavement wears down over time due to traffic polishing, which leads to polished pavement surfaces with lower skid resistance. Three-dimensional (3D) texture parameters can be used to describe the evolution of pavement texture and establish predictive models for skid resistance. In this study, a high-resolution 3D laser scanner and a pendulum friction tester were used to collect 3D texture data and the corresponding friction values of dense-graded asphalt pavement over a period of four years. Fourier transformer and Butterworth filters were applied to decompose the 3D texture data into micro-texture and macro-texture components. Twenty different 3D texture parameters from five categories (height, spatial, hybrid, functional, and feature parameters) were calculated from pavement micro- and macro-textures and optimized using correlation methods to derive an independent set of texture parameters. The performance of a multiple linear regression model and neural network predictive model for predicting skid resistance via selected texture parameters was compared through training and testing. The results indicate that pavement micro-texture contributes more significantly to skid resistance than macro-texture, and neural network models can effectively predict the temporal evolution of skid resistance based on texture data. The neural network model achieves R² values of 0.92 and 0.89 on the training and testing sets, respectively, with RMSE values of 3.37 and 5.45, significantly outperforming the multiple linear regression model (R² = 0.50). Full article

(This article belongs to the Special Issue Remote Sensing for Infrastructure Health Monitoring: Advancements in Sensors and Analysis)

► Show Figures

Figure 1

24 pages, 10558 KB

Open AccessArticle

Hybrid Machine Learning Meta-Model for the Condition Assessment of Urban Underground Pipes

by Mohsen Mohammadagha, Mohammad Najafi, Vinayak Kaushal and Ahmad Jibreen

Infrastructures 2025, 10(11), 282; https://doi.org/10.3390/infrastructures10110282 - 23 Oct 2025

Viewed by 346

Abstract

Urban water infrastructure faces increasing deterioration, necessitating accurate, cost-effective condition assessment. Traditional inspection techniques are intrusive and inefficient, creating demand for scalable machine learning (ML) solutions. This study develops a hybrid ML meta-model to predict underground pipe conditions using a comprehensive dataset of [...] Read more.

Urban water infrastructure faces increasing deterioration, necessitating accurate, cost-effective condition assessment. Traditional inspection techniques are intrusive and inefficient, creating demand for scalable machine learning (ML) solutions. This study develops a hybrid ML meta-model to predict underground pipe conditions using a comprehensive dataset of 11,544 records. The objective is to enhance multi-class classification performance while preserving interpretability. A stacked hybrid architecture was employed, integrating Random Forest, LightGBM, and CatBoost models. Following data preprocessing, feature engineering, and correlation analysis, the neural network-based stacking meta-model achieves 96.67% accuracy, surpassing individual base learners while delivering enhanced robustness through model diversity, improved probability calibration, and consistent performance on challenging intermediate condition classes, which are essential for condition prioritization. Age emerged as the most influential feature, followed by length, material type, and diameter. ROC-AUC scores ranged from 0.894 to 0.998 across all models and classes, confirming high discriminative capability. This work demonstrates hybrid architectures for infrastructure diagnostics. Full article

(This article belongs to the Special Issue Smart Technologies for Sustainable and Resilient Underground Infrastructures)

► Show Figures

Figure 1

21 pages, 5215 KB

Open AccessArticle

Optimization of the Number of Accelerometer Placements for Dynamic Identification of a Historical Masonry Bridge

by Cristiano Giuseppe Coviello, Fabio Rizzo and Maria Francesca Sabbà

Infrastructures 2025, 10(11), 281; https://doi.org/10.3390/infrastructures10110281 - 22 Oct 2025

Viewed by 215

Abstract

Dynamic identification using accelerometers is a common technique for measuring the modal frequencies of existing structures. Strategically placing these sensors on a bridge allows for the derivation of its modal parameters through operational modal analysis (OMA). This study aims to demonstrate how the [...] Read more.

Dynamic identification using accelerometers is a common technique for measuring the modal frequencies of existing structures. Strategically placing these sensors on a bridge allows for the derivation of its modal parameters through operational modal analysis (OMA). This study aims to demonstrate how the number of accelerometers required for the identification of a historical three-arch masonry bridge can be optimized. The experimental campaign involved the Santa Teresa bridge (STb) in Bitonto, a XIX masonry bridge in Southern Italy. Twenty-eight accelerometers were installed on the STb; frequency analysis was first performed with all accelerometers, and then the number was decreased to 13, 8 and 4 accelerometers. The four optimizations performed involved both the number and positioning of accelerometers along the central arch. The five primary vibration modes obtained revealed that with a smaller number of accelerometers, it is possible to correctly identify the natural frequencies of the bridge. A further optimization was per-formed with only No. 6 accelerometers on the keystone of the bridge’s three arches. The results of the modal shapes and natural frequencies showed that a limited number of accelerometers is sufficient to dynamically identify a bridge. The configuration with 13 accelerometers proved to be the best for this purpose. However, optimization with 6 accelerometers proved to be the best for recording normalized displacements compared to the reference configuration. The advantages of this study are directly related to the reduction in time, logistics and costs of in situ monitoring. This preliminary assessment approach enables the establishment of baseline conditions for subsequent periodic monitoring campaigns. Full article

(This article belongs to the Section Infrastructures and Structural Engineering)

► Show Figures

Figure 1

21 pages, 5247 KB

Open AccessArticle

Machine Learning Synthesis of Fire-Following-Earthquake Fragility Surfaces for Steel Moment-Resisting Frames

by Mojtaba Harati and John W. van de Lindt

Infrastructures 2025, 10(11), 280; https://doi.org/10.3390/infrastructures10110280 - 22 Oct 2025

Viewed by 358

Abstract

This paper presents a probabilistic methodology for generating fragility surfaces for low- to mid-rise steel moment-resisting frames (MRFs) under fire-following-earthquake (FFE). The framework integrates nonlinear dynamic seismic analysis, residual deformation transfer, and temperature-dependent fire simulations within a Monte Carlo environment, while explicitly accounting [...] Read more.

This paper presents a probabilistic methodology for generating fragility surfaces for low- to mid-rise steel moment-resisting frames (MRFs) under fire-following-earthquake (FFE). The framework integrates nonlinear dynamic seismic analysis, residual deformation transfer, and temperature-dependent fire simulations within a Monte Carlo environment, while explicitly accounting for uncertainties in structural properties, ground motions, and fire simulation. A fiber-based modeling strategy is employed, combining temperature-sensitive steel materials with fatigue and fracture wrappers to capture cyclic deterioration and abrupt failure. This formulation yields earthquake-only and fire-only fragility curves along the surface boundaries, while interior points quantify the joint fragility response under sequential hazards. The methodology is benchmarked against a machine learning (ML) synthesis framework originally developed for earthquake–tsunami applications and extended here to FFE. Numerical results for a three-story steel MRF show excellent agreement (R² > 0.95, RMSE < 0.02) between simulated and ML-generated surfaces, demonstrating both the efficiency and hazard-neutral adaptability of the ML framework for multi-hazard resilience assessment. Full article

(This article belongs to the Special Issue Earthquake and Multi-Hazard Resilience: Community-Level Insights and AI/ML Applications)

► Show Figures

Figure 1

30 pages, 23419 KB

Open AccessArticle

Leveraging Transformer Models for Seismic Fragility Assessment of Non-Engineered Masonry Structures in Malawi

by Ehsan Harirchian and Viviana Iris Novelli

Infrastructures 2025, 10(11), 279; https://doi.org/10.3390/infrastructures10110279 - 22 Oct 2025

Viewed by 259

Abstract

Assessing seismic vulnerability is a critical step in evaluating the resilience of existing buildings, and fragility curves are widely used to quantify the probability of damage under varying levels of seismic intensity. However, traditional methods for generating these curves often rely on generalized [...] Read more.

Assessing seismic vulnerability is a critical step in evaluating the resilience of existing buildings, and fragility curves are widely used to quantify the probability of damage under varying levels of seismic intensity. However, traditional methods for generating these curves often rely on generalized assumptions that may not accurately capture the seismic behavior of diverse building types within a region. This limitation is particularly evident for non-engineered masonry buildings, which typically lack standardized designs. Their irregular and informal construction makes them difficult to assess using conventional approaches. Transformer-based models, a type of machine learning (ML) technique, offer a promising alternative. These models can identify patterns and relationships in available data, making them well suited for developing seismic fragility curves with improved efficiency and accuracy. While transformers are relatively new to civil engineering, their application to seismic fragility assessment has been largely unexplored. This study presents a pioneering effort to apply transformer models for deriving fragility curves for non-engineered masonry buildings. A comprehensive dataset of 646 masonry buildings observed in Malawi is used to train the models. The transformers are trained to predict the probability of four damage states: Light Damage, Severe Damage, Near Collapse, and Collapse based on Peak Ground Acceleration (PGA). The performance of the transformer-based approach is compared with other ML methods, demonstrating its strong potential for more efficient and accurate seismic fragility assessment. Future work could adopt the proposed methodology and extend the approach by incorporating larger datasets, additional regional contexts, and alternative ML techniques to further enhance predictive performance. Full article

(This article belongs to the Special Issue Earthquake and Multi-Hazard Resilience: Community-Level Insights and AI/ML Applications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Infrastructures, Volume 10, Issue 11 (November 2025) – 14 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI