Feb 2025, Volume 3 Issue 1
    

  • Select all
  • Ankit Kumar, Aditya Parihar

    The study of drainage behavior is essential for using waste material in geotechnical applications. In this study, sandy soil was replaced with waste foundry sand (WFS) at an incremental interval of 20% by weight. Permeability (k) for each mix was acquired at three relative densities (R D), i.e., 65%, 75% and 85%, by using the constant head method. Then the results were further processed with machine learning (ML) models to validate the experimental data. The experimental study demonstrated that k would decrease with the increase in relative density and WFS content. A rise in R D from 65% to 85% resulted in a substantial reduction of up to 140% in the value of k. Moreover, the complete replacement of sand with WFS reduced the value of k by 36%, 51% and 57% for RD of 65%, 75% and 85%, respectively. The total dataset of 90 observations was divided at a ratio of 63/13/15 into training/validation/testing datasets for ML-AI modeling. Input variables include percentage of sand (BS), replacement with WFS, total head (H), time interval (t) and outflow (Q); and k is the output variable. The methods of artificial neural network (ANN), random forest (RF), decision tree (DT) and multi-linear regression (MLR) are used for k prediction. It is found that the random forest approach performed outstandingly in these methods, with an R 2 value of 0.9955. The performance of all the proposed methods was compared and verified with Taylor's diagram. Sensitivity analysis showed that Q and R D were the most influential parameters for predicting k values.

  • Ammu Boban, Preeti Pateriya, Yakshansh Kumar, Kshitij Gaur, Ashutosh Trivedi

    Computer programming-based numerical programs are firmly established in geotechnical engineering, with rapid growth of finite element modeling and machine learning techniques gaining much attention both in practice and academia. This study is intended to expedite the dissemination of advanced computer applications in terms of finite element simulation and machine learning models by investigating the dynamic response of geomaterials subjected to vibratory loads. Several trial models were built to perform the experimental investigations with a vibratory shaker, signal generator, several accelerometers, a data collection system, and other ancillary devices. The implicit integration techniques in commercialized software were adopted for numerical simulations. After data collection from numerical simulation, models were chosen, trained, and assessed to produce predictions that were then used in this study. Several technologies, including the ensemble boosted tree, squared exponential Gaussian Process Regression (GPR), Matern 5/2 GPR, exponential GPR, and decision tree architectures (fine and medium), were used to forecast the displacement of confined geomaterial. The displacement-depth ratio was found rising to 80% in the frequency range of 5 to 25 Hz, suggesting a considerable change in the behavior of the geomaterial. The Matern 5/2 GPR model showed better accuracy with an R2 value of 0.99, indicating an outstanding predictive ability. The Matern 5/2 GPR and boosted tree models could help better understand the links between displacement and its distribution along the direction of load application. The outcomes of this study based on computer-aided finite element programs can be effectively implemented in machine learning to develop computer programs. In conclusion, the computational machine learning models adopted in this study offer a new insight for uncovering hidden intrinsic laws and creating new knowledge for geotechnical researchers and practitioners.

  • Jianguo Cai, Huafei Xu, Jiacheng Chen, Jian Feng, Qian Zhang

    In order to rapidly and accurately evaluate the mechanical properties of a novel origami-inspired tube structure with multiple parameter inputs, this study developed a method of designing origami-inspired braces based on machine learning models. Four geometric parameters, i.e., cross-sectional side length, plate thickness, crease weakening coefficient, and plane angles, were used to establish a mapping relationship with five mechanical parameters, including elastic stiffness, yield load, yield displacement, ultimate load, and ultimate displacement, all of which were calculated from load-displacement curves. Firstly, forward prediction models were trained and compared for single and multiple mechanical outputs. The parameter ranges were extended and refined to improve the predicted results by introducing the intrinsic mechanical relationships. Secondly, certain reverse prediction models were established to obtain the optimized design parameters. Finally, the design method of this study was verified in finite element methods. The design and analysis framework proposed in this study can be used to promote the application of other novel multi-parameter structures.

  • Xi Cheng, Chen Wang, Fayun Liang, Haofen Wang, Xiong Bill Yu

    Underground infrastructure plays a kind of crucial role in modern production and living, especially in big cities where the ground space has been fully utilized. In the context of recent advancements in digital technology, the demand for the application of digital twin technology in underground infrastructure has become increasingly urgent as well. However, the interaction and co-integration between underground engineering entities and virtual models remain relatively limited, primarily due to the unique nature of underground engineering data and the constraints imposed by the development of information technology. This research focuses on underground engineering infrastructure and provides an overview of the application of novel information technologies. Furthermore, a comprehensive framework for digital twin implementation, which encompasses five dimensions and combines emerging technologies, has been proposed. It thereby expands the horizons of the intersection between underground engineering and digital twins. Additionally, a practical project in Wenzhou serves as a case study, where a comprehensive database covering the project’s entire life cycle has been established. The physical model is visualized, endowed with functional implications and data analysis capabilities, and integrated with the visualization platform to enable dynamic operation and maintenance management of the project.

  • Beshoy Maher Hakeem

    In light of rising loads from several sources, including additional stories, eccentric loads, and increased live loads, foundations often face increased demands. To address this, horizontal reinforcements are now commonly positioned beneath footings to enhance the bearing capacity of the loose-dense sand subgrade. By grouting on both sides of the footing, not only can vertical settlement be minimized, but also the soil movement in the horizontal direction under the chosen loaded footing can be reduced. The objective of this study is to conduct extensive experimental work on twenty-one (21) soil models to assess the efficiency of a circular footing resting on granular soil injected with grout diaphragm walls. Specifically, this study investigated the bearing capacity of granular soil in relation to the breadth (b) and length (L) of grouted walls. The results showed that, installing grouted wall injection on both sides of the existing footing is an excellent method to improve the bearing capacity of the subgrade layer. To check the validity of the chosen computational processes, both PLAXIS (3D) software and a 2D Finite Element Program GeoStudio 2018 were used. The findings indicate a direct correlation with the experimental observations in that the reinforcement has a considerable effect on the bearing capacity of a circular-footing resting on granular soil.

  • Darshan Vekaria, Sunil Sinha

    The presence of Artificial Intelligence (AI) and Machine Learning (ML) applications has led to its widespread adoption across diverse domains. AI is making its way into industry, beyond research and academia. Concurrently, the water sector is undergoing a digital transformation. Water utilities in the United States are at different stages in their journey of digital transformation, and the decision makers in water sector, who are non-expert stakeholders in AI applications, need to better understand this technology to make informed decisions. While AI has numerous benefits to offer, there are also many challenges related to data, model development, knowledge integration and ethical concerns that should be considered before implementing it for real world applications. Civil engineering is a licensed profession where critical decision making is involved. Therefore, trust in any decision support technology is critical for its acceptance in real-world applications. Therefore, this research proposes a framework called aiWATERS (Artificial Intelligence for the Water Sector) which can serve as a guide for the water utilities to successfully implement AI in their system. Based on this framework, we conduct pilot interviews and surveys with various small, medium, and large water utilities in the United States (US) to capture their current state of AI implementation and identify the challenges faced by them. The research findings reveal that most of the water utilities in the United States are at an early stage of implementing AI as they face concerns regarding the black box nature, trustworthiness, and sustainability of AI technology in their system. The aiWATERS framework is intended to help the utilities navigate through these issues in their journey of digital transformation.

  • Sumanta Mandal, Amit Shiuly, Debasis Sau, Achintya Kumar Mondal, Kaustav Sarkar

    The construction industry relies so heavily on concrete that it's crucial to precisely forecast and optimize the strength and workability of concrete mixtures, while reducing costs as much as possible. For this objective, this study tries to predict and optimize the compressive strength and workability (slump) of concrete by using deterministic and robust optimization approaches, so as to determine the optimum concrete mixture proportions, while minimizing cost. Specifically, strength and slump were predicted based on concrete mixture proportions with five different machine learning techniques—support vector machine (SVM), artificial neural network (ANN), fuzzy inference system (FIS), adaptive fuzzy inference system (ANIS), and genetic expression programming (GEP), based on a dataset comprising two hundred concrete mixtures, which has various levels of key ingredients, including cement, water, fine aggregate, coarse aggregate, and size of coarse aggregate, along with their associated measures of strength and workability. These ingredients were used as input parameters, while compressive strength and slump (representing workability) served as output parameters for each mix proportion. Experimental investigations were conducted on fifteen distinct concrete mixes to validate the performance of the five networks, finding that ANFIS can yield the best results both for training and validation. This study provides valuable insights for predicting concrete properties and optimizing concrete mixture proportions, thus helping to maximize strength and workability while minimizing costs.

  • Hang Hang Xu, Xue Gang Zhang, Dong Han, Wei Jiang, Yi Zhang, Yu Ming Luo, Xi Hai Ni, Xing Chi Teng, Yi Min Xie, Xin Ren

    Auxetic honeycomb sandwich structures (AHS) composed of a single material generally exhibit comparatively lower energy absorption (EA) and platform stress, as compared to traditional non-auxetic sandwich structures (TNS). To address this limitation, the present study examines the use of aluminum foam (AF) as a filling material in the re-entrant honeycomb sandwich structure (RS). Filling the AHS with AF greatly enhances both the EA and platform stress in comparison to filling the TNS with AF, while the auxetic composite honeycomb sandwich structure effectively addresses interface delamination observed in traditional non-auxetic composite sandwich structures. Subsequently, the positive–negative Poisson’s ratio coupling designs are proposed to strengthen the mechanical features of a single honeycomb sandwich structure. The analysis results show that the coupling structure optimizes the mechanical properties by leveraging the high bearing capacity of the hexagonal honeycomb and the great interaction between the re-entrant honeycomb and the filling material. In contrast with traditional non-auxetic sandwich structures, the proposed auxetic composite honeycomb sandwich structures demonstrate superior EA and platform stress performance, suggesting their immense potential for utilization in protective engineering.

  • Yaping Lai, Yu Li, Yanchen Liu, Peixin Chen, Lijun Zhao, Jin Li, Yi Min Xie

    With rapid advances in design methods and structural analysis techniques, computational generative design strategies have been adopted more widely in the field of architecture and engineering. As a performance-based design technique to find out the most efficient structural form, topology optimization provides a powerful tool for designers to explore lightweight and elegant structures. Building on this background, this study proposes an innovative pedestrian bridge design, which covers the process from conceptualization to detailed design implementation. This pedestrian bridge, with a main span of 152 m, needs to meet some unique architectural requirements, while addressing multiple engineering challenges. Aiming to reduce the depth of the girder but still meeting the load-carrying capacity requirements, the superstructure of this bridge adopts a variable-depth spinal-shaped girder in the center of its deck, thus forming an elegant curving facade, from which one pathway cantilevers on either side. At one end of the bridge, given considerable elevation difference between the bridge deck and the ground, a two-level Fibonacci-type spiral-shaped bicycle ramp is provided. The superstructure is supported by a series of organic tree-shaped branching piers resulting from the topology optimization. The ingenious design for the elegant profile of the bicycle ramp generates an enjoyable and dynamic crossing experience, with scenic views in all directions. By virtue of technological innovation, the pedestrian bridge is expected to create an iconic, cost-effective, and low-maintenance solution. A brief overview of the theoretical background of the bi-directional evolutionary structure optimization (BESO) and the multi-material BESO approach is also offered in this paper, while the construction requirements and challenges, conceptual development process, form-finding strategy, detailed design, and construction method of the bridge are presented.

  • Nur Mohammad Shuman, Mohammad Sadik Khan, Farshad Amini

    Machine learning is frequently used in various geotechnical applications nowadays. This study presents a statistics and machine learning model for settlement prediction of helical piles that relates compressive service load and soil parameters as a group with the pile parameters. Machine learning algorithms such as Decision Trees, Random Forests, AdaBoost, and Artificial Neural Networks (ANN) were used to develop the predictive models. The models were validated using cross-validation techniques and tested on an independent dataset to assess their accuracy and generalizability. Numerical investigation is used here to supplement the field data by simulating various soil conditions and pile geometries that have not been tested in the field. This study compiled numerical results of 3600 models. As the models are well-calibrated and validated, the data from these models can be reasonably assumed to simulate the ground situation. At the end of this study, a comparative analysis of statistic learning and machine learning (ML) was done using the field axial load tests database and numerical investigation on helical piles. It is observed that ML models like Decision Trees and Random Forests provided the better model with R-squared values of 0.92 and 0.96, respectively, for large diameters. The authors believe this study will permit engineers and state agencies to understand this prediction model's efficacy better, resulting in a more resilient approach to designing large-diameter helical piles for the compressive load.

  • Lin Li, Linlong Zuo, Guangfeng Wei, Shouming Jiang, Jian Yu

    Small ground anchors are widely used to fix securing tents in disaster relief efforts. Given the urgent nature of rescue operations, it is crucial to obtain prompt and accurate estimations of their pullout capacity. In this study, a stacking machine learning (ML) model is developed for the rapid estimation of pullout capacity offered by small ground anchors used for temporary tents, leveraging cone penetration data. The proposed stacking model incorporates three ML algorithms as the base regression models: K-nearest neighbors (KNN), support vector regression (SVR), and extreme gradient boosting (XGBoost). A dataset comprising 119 in-situ anchor pullout tests, where the cone penetration data were measured, is utilized to train and assess the stacking model performance. Three metrics, i.e., coefficient of determination (R2), mean absolute error (MAE), and root mean square error (RMSE), are employed to evaluate the predictive accuracy of the proposed model and compare its performance against four popular ML models and an empirical formula to highlight the advantages of the proposed stacking approach. The results affirm that the proposed stacking model outperforms other ML models and the empirical approach as achieving higher R2 and lower MAE and RMSE and more predicted data points falling within 20% error line. Thus, the proposed stacking model holds promising potential as a solution for efficiently predicting the pullout capacity of small ground anchors.

  • Tao Xu, Xiaodong Huang, Xiaoshan Lin, Yi Min Xie

    Topology optimization techniques are increasingly utilized in structural design to create efficient and aesthetically pleasing structures while minimizing material usage. Many existing topology optimization methods may generate slender structural members under compression, leading to significant buckling issues. Consequently, incorporating buckling considerations is essential to ensure structural stability. This study investigates the capabilities of the bi-directional evolutionary structural optimization method, particularly its extension to handle multiple load cases in buckling optimization problems. The numerical examples presented focus on three classical cases relevant to civil engineering: maximizing the buckling load factor of a compressed column, performing buckling-constrained optimization of a frame structure, and enhancing the buckling resistance of a high-rise building. The findings demonstrate that the algorithm can significantly improve structural stability with only a marginal increase in compliance. The detailed mathematical modeling, sensitivity analyses, and optimization procedures discussed provide valuable insights and tools for engineers to design structures with enhanced stability and efficiency.

  • John Igeimokhia Braimah, Wasiu Olabamiji Ajagbe, Kolawole Adisa Olonade

    Quarry dust, conventionally considered waste, has emerged as a potential solution for sustainable construction materials. This paper comprehensively review the mechanical properties of blocks manufactured from quarry dust, with a particular focus on the transformative role of machine learning (ML) in predicting and optimizing these properties. By systematically reviewing existing literature and case studies, this paper evaluates the efficacy of ML methodologies, addressing challenges related to data quality, feature selection, and model optimization. It underscores how ML can enhance accuracy in predicting mechanical properties, providing a valuable tool for engineers and researchers to optimize the design and composition of blocks made from quarry dust. This synthesis of mechanical properties and ML applications contributes to advancing sustainable construction practices, offering insights into the future integration of technology for predictive modeling in material science.

  • Sesugh Terlumun, M. E. Onyia, F. O. Okafor

    Concrete is one of the most common construction materials used all over the world. Estimating the strength properties of concrete traditionally demands extensive laboratory experimentation. However, researchers have increasingly turned to predictive models to streamline this process. This review focuses on predicting the compressive strength of self-compacting concrete using artificial intelligence (AI) techniques. Self-compacting concrete represents an advanced construction material particularly suited for scenarios where traditional vibrational methods face limitations due to intricate formwork or reinforcement complexities. This review evaluates various AI techniques through a comparative performance analysis. The findings highlight that employing Deep Neural Network models with multiple hidden layers significantly enhances predictive accuracy. Specifically, artificial neural network (ANN) models exhibit robustness, consistently achieving R2 values exceeding 0.7 across reviewed studies, thereby demonstrating their efficacy in predicting concrete compressive strength. The integration of ANN models is recommended for formulating various civil engineering properties requiring predictive capabilities. Notably, the adoption of AI models reduces both time and resource expenditures by obviating the need for extensive experimental testing, which can otherwise delay construction activities.

  • Furkan Kizilay, Mina R. Narman, Hwapyeong Song, Husnu S. Narman, Cumhur Cosgun, Ammar Alzarrad

    Earthquakes pose a significant threat to life and property worldwide. Rapid and accurate assessment of earthquake damage is crucial for effective disaster response efforts. This study investigates the feasibility of employing deep learning models for damage detection using drone imagery. We explore the adaptation of models like VGG16 for object detection through transfer learning and compare their performance to established object detection architectures like YOLOv8 (You Only Look Once) and Detectron2. Our evaluation, based on various metrics including mAP, mAP50, and recall, demonstrates the superior performance of YOLOv8 in detecting damaged buildings within drone imagery, particularly for cases with moderate bounding box overlap. This finding suggests its potential suitability for real-world applications due to the balance between accuracy and efficiency. Furthermore, to enhance real-world feasibility, we explore two strategies for enabling the simultaneous operation of multiple deep learning models for video processing: frame splitting and threading. In addition, we optimize model size and computational complexity to facilitate real-time processing on resource-constrained platforms, such as drones. This work contributes to the field of earthquake damage detection by (1) demonstrating the effectiveness of deep learning models, including adapted architectures, for damage detection from drone imagery, (2) highlighting the importance of evaluation metrics like mAP50 for tasks with moderate bounding box overlap requirements, and (3) proposing methods for ensemble model processing and model optimization to enhance real-world feasibility. The potential for real-time damage assessment using drone-based deep learning models offers significant advantages for disaster response by enabling rapid information gathering to support resource allocation, rescue efforts, and recovery operations in the aftermath of earthquakes.

  • M. Lalitha Pallavi, Subhashish Dey, Ganugula Taraka Naga Veerendra, Siva Shanmukha Anjaneya Babu Padavala, Akula Venkata Phani Manoj

    The yearly production of plastic garbage is rising in the current environment as a result of the fast population rise. Recycling and reusing plastic trash is essential for sustainable development. The need of the hour is to utilize waste polythene for various supporting reasons since it is not biodegradable. These materials are made of polymers like polyethylene, polypropylene, and polystyrene. Due to the enhanced performance and elimination of the environmental issue, adding plastic waste to flexible pavement has emerged as a desirable choice. A composite material known as bituminous concrete (BC) is often utilized in construction projects such as road paving, airport terminals, and stopover areas. It includes mineral aggregate and black top or bitumen, which are combined, laid down in layers, and then compacted. The bituminous mixture in this research article was combined with plastic to use a chemical stabilizer. The ideal bitumen content is replaced by 0, 15%, 27%, and 36% plastic, as well as the bitumen's weight, stability, and Marshall value to create hypothermal. A linear scale is used to compare the flow rates to the bituminous mixture. The characterization of plastics contains bituminous materials are done by the SEM–EDX, XRD, FTIR and BET analysis. There have been several studies on the addition of trash to bituminous mixes, but this one is focused on the use of plastic waste as a modification in a bitumen binder for flexible pavement. According to research, bituminous mixes containing up to 4 percent plastic waste are excellent for sustainable development.

  • Saeedeh Ghaemifard, Amin Ghannadiasl

    Optimization is the process of creating the best possible outcome while taking into consideration the given conditions. The ultimate goal of optimization is to maximize or minimize the desired effects to meet the technological and management requirements. When faced with a problem that has several possible solutions, an optimization technique is used to identify the best one. This involves checking different search domains at the right time, depending on the specific problem. To solve these optimization problems, nature-inspired algorithms are used as part of stochastic methods. In civil engineering, numerous design optimization problems are nonlinear and can be difficult to solve via traditional techniques. In such points, metaheuristic algorithms can be a more useful and practical option for civil engineering usages. These algorithms combine randomness and decisive paths to compare multiple solutions and select the most satisfactory one. This article briefly presents and discusses the application and efficiency of various metaheuristic algorithms in civil engineering topics.

  • Rashid Mustafa

    This study is primarily aimed at creating three machine learning models: artificial neural network (ANN), random forest (RF), and k-nearest neighbour (KNN), so as to predict the crippling load (CL) of I-shaped steel columns. Five input parameters, namely length of column (L), width of flange (b f), flange thickness (t f), web thickness (t w) and height of column (H), are used to compute the crippling load (CL). A range of performance indicators, including the coefficient of determination (R 2), variance account factor (VAF), a-10 index, root mean square error (RMSE), mean absolute error (MAE) and mean absolute deviation (MAD), are used to assess the effectiveness of the established machine learning models. The results show that all of the three ML (machine learning) models can accurately predict the crippling load, but the performance of ANN is superior: it delivers the highest value of R 2 = 0.998 and the lowest value of RMSE = 0.008 in the training phase, as well as the highest value of R 2 = 0.996 and the smaller value of RMSE = 0.012 in the testing phase. Additional methods, including rank analysis, reliability analysis, regression plot, Taylor diagram and error matrix plot, are employed to assess the models’ performance. The reliability index (β) of the models is calculated by using the first-order second moment (FOSM) technique, and the result is compared with the actual value. Additionally, sensitivity analysis is performed to check the impact of the input variables on the output (CL), finding that b f has the greatest impact on the crippling load, followed by t f, t w, H and L, in that order. This study demonstrates that ML techniques are useful for developing a reliable numerical tool for measuring the crippling load of I-shaped steel columns. It is found that the proposed techniques can also be used to predict other kinds of failures as well as different kinds of perforated columns.

  • Umar Jibrin Muhammad, Ismail I. Aminu, Ismail A. Mahmoud, U. U. Aliyu, A. G. Usman, Mahmud M. Jibril, Salim Idris Malami, Sani I. Abba

    Traditional methods for proportioning of high-performance concrete (HPC) have certain shortcomings, such as high costs, usage constraints, and nonlinear relationships. Implementing a strategy to optimize the mixtures of HPC can minimize design expenses, time spent, and material wastage in the construction sector. Due to HPC's exceptional qualities, such as high strength (HS), fluidity and resilience, it has been broadly used in construction projects. In this study, we employed Generalized Regression Neural Network (GRNN), Nonlinear AutoRegressive with exogenous inputs (NARX neural network), and Random Forest (RF) models to estimate the Compressive Strength (CS) of HPC in the first scenario. In contrast, the second scenario involved the development of an ensemble model using the Radial Basis Function Neural Network (RBFNN) to detect inferior performance of standalone model combinations. The output variable was the 28 Days CS in MPa, while the input variables included slump (S), water-binder ratio (W/B) %, water content (W) kg/m3, fine aggregate ratio (S/a) %, silica fume (SF)%, and superplasticizer (SP) kg/m3. An RF model was developed by using R Studio; GRNN and NARX-NN models were developed by using the MATLAB 2019a toolkit; and the pre- and post-processing of data was carried out by using E-Views 12.0. The results indicate that in the first scenario, the Combination M1 of the RF model outperformed other models, with greater prediction accuracy, yielding a PCC of 0.854 and MAPE of 4.349 during the calibration phase. In the second scenario, the ensemble of RF models surpassed all other models, achieving a PCC of 0.961 and MAPE of 0.952 during the calibration phase. Overall, the proposed models demonstrate significant value in predicting the CS of HPC.

  • Mustapha A. Raji, Boluwatife M. Falola, Jesse T. Enikuomehin, Akintoye O. Oyelade, Yetunde O. Abiodun, Yusuf A. Olaniyi, Olusola G. Olagunju, Kosisochukwu L. Anyaegbuna, Musa O. Abdulkareem, Christopher A. Fapohunda

    The use of agricultural by-products, such as Rice Husk Ash (RHA), in concrete production has gained significant attention as a sustainable alternative to traditional construction materials. This study aims to evaluate and compare the effects of Nano-Rice Husk Ash (NRHA) and Micro-Rice Husk Ash (MRHA) on the compressive strength of concrete. Concrete samples were prepared with varying replacement levels of NRHA (0% to 3%) and MRHA (0% to 14%) and underwent thorough examination through both slump and compressive strength tests conducted at 7, 21, 28, and 56 days. The results showed that NRHA achieved maximum compressive strength at a 1% replacement level, while MRHA reached its peak at a 0.5% replacement level. However, a comparison of the compressive strength of NRHA at 1% (22 N/mm2) against MRHA at 0.5% (21.5 N/mm2) revealed that the marginal difference in strength made MRHA a more cost-effective option due to the lower expenses involved in its preparation. Thus, MRHA presents a more economical solution for achieving comparable compressive strength. Furthermore, the study applied linear, non-linear, and mixed regression analyses to model the properties of NRHA and MRHA concrete based on a comprehensive set of variables. The analysis found that the blended ordinary and logarithmic models provided the best fit, offering superior accuracy compared to linear and non-linear models.

  • Tatiana Fountoukidou, Iuliia Tkachenko, Benjamin Poli, Serge Miguet

    A number of bridges have collapsed around the world over the past years, with detrimental consequences on safety and traffic. To a large extend, such failures can be prevented by regular bridge inspections and maintenance, tasks that fall in the general category of structural health monitoring (SHM). Those procedures are time and labor consuming, which partly accounts for their neglect. Computer vision and artificial intelligence (AI) methods have the potential to ease this burden, by fully or partially automating bridge monitoring. A critical step in this automation is the identification of a bridge’s structural components. In this work, we propose an extensible synthetic dataset for structural component semantic segmentation of portal frame bridges (PFBridge). We first create a 3 dimensional (3D) generic mesh representing the bridge geometry, while respecting a set of rules. The definition of new, or the extension of the existing rules can adjust the dataset to specific needs. We then add textures and other realistic elements to the model, and create an automatically annotated synthetic dataset. The synthetic dataset is used in order to train a deep semantic segmentation model to identify bridge components on bridge images. The amount of available real images is not sufficient to entirely train such a model, but is used to refined the model trained on the synthetic data. We evaluate the contribution of the dataset to semantic segmentation by training several segmentation models on almost 2,000 synthetic images and then finetuning with 88 real images. The results show an increase of 28% on the F1-score when the synthetic dataset is used. To demonstrate a potential use case, the model is integrated in a 3D point cloud capturing system, producing an annotated point cloud where each point is associated with a semantic category (structural component). Such a point cloud can then be used in order to facilitate the generation of a bridge’s digital twin.

  • Jiapei Zhao, Nobuyoshi Yabuki, Tomohiro Fukuda

    Effective water management and flood prevention are critical challenges encountered by both urban and rural areas, necessitating precise and prompt monitoring of waterbodies. As a fundamental step in the monitoring process, waterbody segmentation involves precisely delineating waterbody boundaries from imagery. Previous research using satellite images often lacks the resolution and contextual detail needed for local-scale analysis. In response to these challenges, this study seeks to address them by leveraging common natural images that are more easily accessible and provide higher resolution and more contextual information compared to satellite images. However, the segmentation of waterbodies from ordinary images faces several obstacles, including variations in lighting, occlusions from objects like trees and buildings, and reflections on the water surface, all of which can mislead algorithms. Additionally, the diverse shapes and textures of waterbodies, alongside complex backgrounds, further complicate this task. While large-scale vision models have typically been leveraged for their generalizability across various downstream tasks that are pre-trained on large datasets, their application to waterbody segmentation from ground-level images remains underexplored. Hence, this research proposed the Visual Aquatic Generalist (VAGen) as a countermeasure. Being a lightweight model for waterbody segmentation inspired by visual In-Context Learning (ICL) and Visual Prompting (VP), VAGen refines large visual models by innovatively adding learnable perturbations to enhance the quality of prompts in ICL. As demonstrated by the experimental results, VAGen demonstrated a significant increase in the mean Intersection over Union (mIoU) metric, showing a 22.38% enhancement when compared to the baseline model that lacked the integration of learnable prompts. Moreover, VAGen surpassed the current state-of-the-art (SOTA) task-specific models designed for waterbody segmentation by 6.20%. The performance evaluation and analysis of VAGen indicated its capacity to substantially reduce the number of trainable parameters and computational overhead, and proved its feasibility to be deployed on cost-limited devices including unmanned aerial vehicles (UAVs) and mobile computing platforms. This study thereby makes a valuable contribution to the field of computer vision, offering practical solutions for engineering applications related to urban flood monitoring, agricultural water resource management, and environmental conservation efforts.