ADVANCING SUSTAINABLE WATER MANAGEMENT THROUGH CIVIL ENGINEERING INNOVATION

Sustainable water management is a fundamental pillar of global environmental sustainability and resource conservation. With increasing water demand driven by climate change, rapid urbanization, and population growth, the need for innovative and efficient water management strategies has become more urgent than ever. Civil engineering plays a central role in designing, implementing, and optimizing systems that ensure reliable water supply, effective wastewater treatment, and resilient stormwater management. This study critically evaluates how modern engineering approaches contribute to achieving long-term sustainability goals.

Innovations in Water Supply Systems

Recent advancements in water supply systems focus on improving efficiency, reducing losses, and ensuring water quality. Technologies such as smart monitoring systems, leak detection networks, advanced filtration, and desalination are transforming how water is sourced and distributed. Civil engineers are integrating digital tools and data-driven approaches to optimize system performance, minimize wastage, and enhance resilience against climate variability, ensuring sustainable and continuous water availability.

Transformations in Wastewater Treatment

Wastewater treatment has evolved from simple disposal systems to resource recovery platforms. Modern treatment technologies emphasize energy efficiency, nutrient recovery, and water reuse. Biological treatment processes, membrane technologies, and decentralized treatment systems are enabling the recycling of wastewater for agricultural and industrial use. These innovations not only reduce environmental pollution but also contribute to circular economy practices by turning waste into valuable resources.

Sustainable Stormwater Management Strategies

Stormwater management has shifted toward sustainable and nature-based solutions that reduce flooding risks and enhance urban resilience. Green infrastructure approaches, such as permeable pavements, rain gardens, bioswales, and retention systems, help manage runoff while improving groundwater recharge and urban ecosystems. These strategies support climate adaptation and reduce pressure on conventional drainage systems, making cities more resilient to extreme weather events.

Challenges in Implementation and Adoption

Despite technological progress, several barriers hinder the widespread adoption of sustainable water management solutions. Financial constraints, regulatory limitations, technological gaps, and lack of public awareness remain significant challenges. Additionally, integrating new technologies into existing infrastructure requires careful planning and investment. Addressing these challenges requires coordinated efforts among policymakers, engineers, and stakeholders to create supportive frameworks for innovation.

Strategic Pathways for Sustainable Water Management

To overcome these challenges, the study proposes strategic approaches including policy reform, investment in advanced technologies, public–private partnerships, and capacity building. Integrating theoretical frameworks with real-world case studies provides practical insights into effective implementation. By promoting interdisciplinary collaboration and encouraging innovation, civil engineering can lead the transition toward sustainable, efficient, and equitable water resource management systems that meet current and future demands.

#WaterConservation

#EnvironmentalEngineering

#SmartWater

#CircularEconomy

#SustainableCities

#InfrastructureInnovation

#WaterSustainability

#ClimateAdaptation

#EngineeringSolutions

#FutureInfrastructure

#GlobalSustainability

#CivilEngineeringResearch

ERGONOMICS-BASED MANAGEMENT STRATEGIES FOR HIGH-ALTITUDE TUNNEL CONSTRUCTION

High-altitude tunnel construction presents unique physiological and operational challenges due to reduced oxygen availability, extreme environmental conditions, and increased physical strain on workers. These factors significantly elevate construction risks and can negatively impact productivity. This study investigates ergonomics-based construction management strategies aimed at improving worker safety and efficiency by analyzing respiratory metabolism and energy consumption under high-altitude conditions.

Impact of High Altitude on Worker Physiology

Field testing and controlled bicycle power simulation experiments were conducted to evaluate the effects of altitude on respiratory metabolic parameters and energy metabolism rate (EMR). Results indicate that as altitude increases, oxygen intake efficiency decreases, leading to higher physiological stress and energy expenditure. Workers operating in high-altitude environments must therefore exert greater effort to perform the same tasks compared to those at lower elevations.

Energy Metabolism Rate Variation with Altitude

The study reveals a significant increase in EMR with rising altitude. Specifically, EMR values increased from 8–11 kJ/(min·m²) at 2500 m to 10–14 kJ/(min·m²) at 4700 m, indicating a substantial rise in energy demand. The altitude range of 3400–3800 m was identified as a critical physiological adaptation zone, where workers begin to experience noticeable metabolic strain. At elevations above 4000 m, task-specific metabolic rates (MET) vary significantly, highlighting the growing influence of individual physiological differences.

Development of Work Duration Control Standards

Based on observed metabolic changes, the study establishes scientifically grounded work duration control standards for high-altitude tunnel construction. These standards are designed to prevent excessive fatigue, reduce health risks, and maintain consistent productivity. By aligning work-rest cycles with physiological limits, construction managers can better protect workers from altitude-related stress and performance decline.

Targeted Oxygen Supply Strategy

A novel “3 regions + 5 oxygen supply measures” strategy was proposed to optimize oxygen delivery in high-altitude tunnel environments. This approach categorizes work zones based on oxygen demand and implements targeted oxygen supply interventions tailored to each region. The strategy ensures efficient oxygen utilization, reduces unnecessary resource consumption, and enhances worker adaptability to altitude conditions.

Performance Improvements and Practical Implications

Implementation of the proposed ergonomics-based management strategies resulted in significant improvements. During a six-week monitoring period, worker efficiency increased by 13.6% to 28.6%, while cases requiring medical treatment due to oxygen deficiency dropped from 10–15 cases per week to zero. These outcomes demonstrate the effectiveness of integrating physiological insights into construction management practices. The findings provide a strong theoretical and practical foundation for improving safety and efficiency in high-altitude tunnel projects and can be extended to other high-altitude engineering applications.

#OccupationalHealth

#SmartConstruction

#ProductivityImprovement

#HumanFactorsEngineering

#EngineeringInnovation

#SustainableConstruction

#WorkplaceSafety

#HighAltitudeWork

#CivilEngineeringResearch

#ConstructionEfficiency

AGMAMBA: A LIGHTWEIGHT ADAPTIVE GUIDED STATE SPACE MODEL FOR PIXEL-LEVEL CRACK SEGMENTATION IN CIVIL INFRASTRUCTURE

Accurate pixel-level segmentation of crack regions is essential for the inspection and maintenance of civil infrastructure such as bridges, pavements, and buildings. Early detection of structural cracks enables timely maintenance actions and reduces the risk of structural failure. However, existing segmentation approaches often struggle to simultaneously capture fine crack textures and suppress complex background noise, which can significantly degrade detection accuracy. Additionally, many high-performing models require large computational resources, limiting their applicability in real-time infrastructure monitoring systems.

Challenges in Existing Crack Segmentation Methods

Current deep learning-based crack detection methods frequently encounter two primary challenges. First, cracks typically exhibit thin, irregular, and discontinuous patterns, making it difficult for models to dynamically capture their texture and morphology. Second, background elements such as shadows, stains, and surface roughness can easily be misidentified as cracks. These challenges often lead to reduced segmentation accuracy or excessive computational complexity when attempting to improve feature extraction.

AGMamba Architecture and Lightweight Design

To overcome these limitations, this study introduces AGMamba, a lightweight adaptive guided state space model designed for efficient crack segmentation. The architecture focuses on capturing crack texture features while minimizing redundant background information. With only 3.40 million parameters and 20.99 GFLOPs, AGMamba achieves a strong balance between computational efficiency and segmentation performance, making it suitable for practical civil infrastructure inspection applications.

Crack Perception Module (CPM)

A key component of the proposed model is the Crack Perception Module (CPM), which integrates two complementary mechanisms:

• Adaptive Guided Scanning Strategy (AGSS2D) – prioritizes crack regions during feature scanning to improve the efficiency of texture extraction.

• Selective Key Clue Modeling (SKCM) – selectively aggregates information from critical crack edges and structural features.

Together, these modules allow the network to focus on meaningful crack patterns while reducing the influence of irrelevant background features.

Frequency-Domain Feature Perception

To further enhance segmentation performance, the model incorporates a High-Low Frequency Feature Perception (HLFP) strategy and a Frequency-Domain Segmentation Head (FDSH). These components analyze differences between high-frequency crack textures and low-frequency background patterns. By leveraging frequency-domain information, the framework effectively suppresses background interference and improves crack boundary detection accuracy.

Experimental Results and Performance Evaluation

Extensive experiments were conducted on four public crack datasets, demonstrating that AGMamba consistently outperforms existing state-of-the-art (SOTA) segmentation models. On the Crack500 dataset, the proposed model achieved an F1 score of 0.7622 and an mIoU of 0.7808, representing improvements of 1.41% and 1.21%, respectively, over previous SOTA methods. These results confirm the model’s ability to achieve high segmentation accuracy while maintaining low computational cost, making it highly suitable for automated infrastructure inspection systems.

Global Civil Engineering Awards

🔗 Nominate now! 👉 https://civilengineeringawards.com/award-nomination/?ecategory=Awards&rcategory=Awardee

🌐 Visit: civilengineeringawards.com

📩 Contact: contact@civilengineeringawards.com

#BridgeInspection
#PavementMonitoring
#AIinCivilEngineering
#StructuralSafety
#AutomatedInspection
#DigitalInfrastructure
#LightweightAI
#EngineeringVision
#CivilEngineeringResearch
#SmartMaintenance

 

HYBRID PKO-XGBOOST MODEL FOR ACCURATE SHEAR STRENGTH PREDICTION OF CONCRETE-FILLED STEEL TUBE COLUMNS

Concrete-filled steel tubes (CFST) are extensively used in modern civil engineering structures because of their excellent load-bearing capacity, ductility, and seismic resistance. The composite interaction between the steel tube and the concrete core significantly enhances structural performance compared to conventional steel or reinforced concrete members. However, existing design standards, including the American Institute of Steel Construction (AISC) and Eurocode 4 provisions, often provide conservative shear strength predictions because they do not fully capture the complex composite mechanisms governing CFST behavior.

Limitations of Conventional Design Methods

Traditional empirical or semi-empirical design equations rely on simplified assumptions about material interaction and stress distribution. As a result, these models often underestimate the actual shear capacity of CFST columns, leading to conservative structural designs and inefficient material usage. Furthermore, such approaches cannot easily adapt to large experimental datasets or capture nonlinear relationships among structural parameters. This limitation motivates the adoption of data-driven predictive models capable of learning complex interactions within structural systems.

Hybrid Machine Learning Framework

To improve prediction accuracy, this study proposes a hybrid modeling approach that integrates Extreme Gradient Boosting (XGBoost) with the Pied Kingfisher Optimizer (PKO). PKO is a nature-inspired optimization algorithm designed to enhance model performance by efficiently tuning hyperparameters. By combining the strong learning capability of XGBoost with PKO's optimization strategy, the hybrid model achieves improved predictive capability and robustness when estimating the shear strength of CFST columns.

Prediction Interval and Error Mitigation Techniques

Beyond point predictions, the study incorporates quantile regression to generate prediction intervals for ultimate shear force, enabling uncertainty quantification in structural predictions. Additionally, the Asymmetric Squared Error Loss (ASEL) function is introduced to reduce the risk of overestimation errors, which are critical in structural safety evaluations. This approach ensures that the predictive model remains both accurate and conservative where necessary, aligning with engineering safety requirements.

Model Performance and Comparative Analysis

Computational results demonstrate that the PKO-XGBoost model significantly outperforms conventional models. The hybrid framework achieves a Mean Absolute Percentage Error (MAPE) of 4.431% and a coefficient of determination (R²) of 0.9925 on the test dataset. Furthermore, the ASEL-PKO-XGBoost variant effectively reduces overestimation errors to 28.26% while maintaining comparable predictive performance. These results confirm the effectiveness of the proposed framework for accurately predicting CFST shear capacity.

Development of Predictive Equation and Practical Tools

In addition to the machine learning model, a new strength equation was derived using a Genetic Algorithm (GA) combined with existing equation-based models. The resulting equation demonstrates improved accuracy (R² = 0.934) compared with traditional design formulas. To facilitate practical implementation, web-based Graphical User Interfaces (GUIs) were also developed, enabling engineers to perform real-time shear strength predictions efficiently. These tools support practical adoption of advanced predictive methods in structural design and engineering practice.

Global Civil Engineering Awards

🔗 Nominate now! 👉 https://civilengineeringawards.com/award-nomination/?ecategory=Awards&rcategory=Awardee

🌐 Visit: civilengineeringawards.com

📩 Contact: contact@civilengineeringawards.com

#SmartStructuralDesign

#AIinConstruction

#GeneticAlgorithm

#EngineeringPrediction

#StructuralSafety

#DataDrivenEngineering

#SteelConcreteComposite

#EngineeringInnovation

#AdvancedStructuralModels

#DigitalEngineering

 

MICROWAVE SINTERING OF ENGINEERING SPOIL CERAMSITE: MULTI-PHYSICS MODELING AND TEMPERATURE FIELD OPTIMIZATION

Microwave sintering of engineering spoil into ceramsite presents a sustainable alternative to traditional high-temperature kiln processes. By converting construction and excavation waste into lightweight aggregate, this technology supports resource recycling and environmentally responsible construction practices. However, challenges such as temperature nonuniformity and inefficient energy utilization limit its large-scale industrial application. This study addresses these challenges by developing a comprehensive multi-physics modeling framework to analyze the energy conversion process and temperature evolution during microwave sintering.

Multi-Physics Coupled Modeling Framework

A three-dimensional electromagnetic–thermal–radiation coupled model was developed to simulate the microwave sintering process of ceramsite. The model integrates electromagnetic wave propagation, thermal conduction, and radiative heat transfer mechanisms to capture the complex interactions occurring during heating. Experimental validation confirmed the model’s accuracy in predicting temperature distribution and heating behavior, providing a reliable tool for investigating energy transfer and temperature field development in microwave-assisted sintering systems.

Role of Silicon Carbide Susceptor in Hybrid Heating

The introduction of a silicon carbide (SiC) susceptor significantly improves heating efficiency and temperature uniformity. Due to the relatively low dielectric loss of ceramsite materials, direct microwave absorption is limited. The SiC susceptor acts as an auxiliary heating element, converting microwave energy into thermal energy and transferring heat to surrounding ceramsite particles. This hybrid heating mechanism reduces temperature gradients and enhances overall sintering stability.

Temperature Evolution and Heat Transfer Mechanisms

The temperature field evolves through different dominant heat transfer mechanisms during the sintering process. At lower temperatures (below 800 °C), localized microwave-induced hotspots result in uneven heating patterns. As the temperature increases beyond 1000 °C, radiative heat transfer becomes the primary mechanism, promoting more uniform temperature distribution across the material. This transition highlights the importance of considering both electromagnetic and radiative effects when designing microwave sintering systems.

Influence of Particle Size and Microwave Power

Parametric analysis revealed that ceramsite particle size and microwave power significantly influence heating uniformity and energy efficiency. Smaller particles (1 cm) produce more uniform temperature distributions, while larger particles (3 cm) are susceptible to uneven heating due to electric field intensity variations. Regarding power input, lower microwave power improves temperature uniformity but increases energy consumption, whereas higher power reduces energy usage but worsens temperature gradients. A moderate microwave power of 3 kW was identified as the optimal operating condition for balancing energy efficiency and thermal uniformity.

Scale-Up Strategies and Hotspot Mitigation

To address industrial-scale challenges, the study proposed an enclosed susceptor design with multi-layer ceramsite arrangements. Among tested configurations, a two-layer structure achieved optimal heat exchange, reducing the temperature coefficient of variation (COV-T) by 21.1% compared to a single-layer setup. Additionally, rotational heating of the susceptor was introduced to mitigate hotspot formation. This dynamic heat redistribution mechanism significantly improves temperature uniformity, achieving a COV-T value of 0.014 at 1240 °C. Thermal flux analysis indicates that alternating radiative heat exchange between the rotating susceptor and ceramsite particles is the key mechanism behind the enhanced uniformity.

Global Civil Engineering Awards

🔗 Nominate now! 👉 https://civilengineeringawards.com/award-nomination/?ecategory=Awards&rcategory=Awardee

🌐 Visit: civilengineeringawards.com

📩 Contact: contact@civilengineeringawards.com

#MaterialsEngineering

#CircularEconomy

#LightweightAggregates

#EnergyEfficiency

#CivilEngineeringResearch

#IndustrialSintering

#GreenMaterials

#ConstructionInnovation

#ThermalModeling

MULTIFUNCTIONAL AUXETIC METAMATERIALS: DESIGN STRATEGIES AND MULTIPHYSICS COUPLING MECHANISMS

Auxetic metamaterials, characterized by their negative Poisson’s ratio, represent a rapidly evolving class of engineered materials with unconventional mechanical behavior. Unlike traditional materials that contract laterally when stretched, auxetic structures expand laterally under tensile loading due to their unique geometric deformation mechanisms. This property enables exceptional strain energy redistribution, resulting in enhanced stiffness, strength, and energy absorption capabilities. In recent years, auxetic metamaterials have expanded beyond purely mechanical applications and are increasingly recognized as multifunctional platforms capable of interacting with thermal, acoustic, electrical, magnetic, and optical fields.

Deformation Mechanisms and Structural Behavior

The exceptional performance of auxetic metamaterials originates from their distinct deformation mechanisms, including rotational units, re-entrant cell structures, and chiral geometries. These structural configurations allow strain energy to concentrate and redistribute efficiently within the material, improving resistance to fracture and impact. Such deformation characteristics contribute to higher mechanical stability, superior toughness, and improved energy dissipation compared to conventional materials. Understanding these deformation principles is fundamental to tailoring auxetic materials for advanced engineering applications.

Design Strategies for Multifunctional Auxetic Structures

Recent advances in auxetic metamaterial design rely on innovative strategies that enhance structural performance and enable multifunctional behavior. Key design approaches include geometric reconfiguration of unit cells, instability engineering to trigger controlled deformation modes, hierarchical structuring to combine properties across multiple scales, and multi-material integration to introduce additional functional capabilities. These strategies enable the development of programmable materials whose mechanical responses can be customized for specific engineering requirements.

Multiphysics Coupling and Functional Capabilities

Auxetic metamaterials demonstrate strong interactions with external physical fields, enabling multiphysics coupling effects. Their structural configuration allows them to regulate thermal expansion, absorb and dissipate mechanical impact energy, and manipulate acoustic and elastic wave propagation across wide frequency ranges. These capabilities create opportunities for adaptive materials that can respond dynamically to environmental stimuli, making auxetic structures suitable for advanced sensing, vibration mitigation, and thermal management applications.

Emerging Applications Across Engineering Fields

The multifunctional characteristics of auxetic metamaterials have opened pathways for innovative applications across multiple engineering disciplines. In biomedical engineering, auxetic structures offer improved compatibility with biological tissues and enhanced implant performance. Aerospace and civil infrastructure benefit from their energy absorption and vibration control properties, while soft robotics utilizes their flexible yet resilient structural behavior for adaptive motion systems. These diverse applications highlight the transformative potential of auxetic metamaterials in next-generation technologies.

Challenges and Future Research Directions

Despite significant progress, several challenges remain before auxetic metamaterials can be widely implemented in practical systems. Scalable manufacturing methods, long-term durability under coupled mechanical and environmental loading, and reliable predictive multiphysics modeling are key areas requiring further research. Future advancements are expected to emerge from the integration of topology optimization, advanced fabrication technologies such as additive manufacturing, and system-level experimental validation. Addressing these challenges will accelerate the transition of multifunctional auxetic metamaterials from laboratory concepts to real-world engineering solutions.

Global Civil Engineering Awards

🔗 Nominate now! 👉 https://civilengineeringawards.com/award-nomination/?ecategory=Awards&rcategory=Awardee

🌐 Visit: civilengineeringawards.com

📩 Contact: contact@civilengineeringawards.com

#AdditiveManufacturing

#AcousticMetamaterials

#ThermalEngineering

#WavePropagation

#EnergyAbsorption

#MaterialScienceResearch

#BiomedicalEngineering

#AerospaceMaterials

#CivilEngineeringInnovation

#SoftRobotics

#NextGenMaterials

 

GEOMETRIC FEATURE–BASED IDENTIFICATION OF ROCK MASS DISCONTINUITIES USING ADAPTIVE HIERARCHICAL CLUSTERING

Discontinuities such as joints, fractures, and bedding planes play a decisive role in controlling the stability and safety of underground engineering structures. Accurate identification and grouping of these discontinuities are essential for tunnel excavation, slope stabilization, and foundation design. Conventional identification methods primarily depend on normal vector estimation combined with directional clustering algorithms. However, these approaches often suffer from reduced accuracy under uneven orientation density, omission of critical discontinuities, and the need for manual parameter tuning. To address these limitations, this study proposes a novel discontinuity identification framework based on geometric feature analysis and adaptive statistical clustering.

Limitations of Conventional Directional Clustering Methods

Traditional approaches rely heavily on estimating surface normals from point cloud data and grouping them via clustering techniques such as k-means or density-based methods. These strategies encounter three key challenges:

(1) performance degradation when discontinuity orientations are unevenly distributed,

(2) loss of small or sparse structural sets, and

(3) dependence on manually predefined cluster numbers or thresholds.

Such limitations reduce reliability in complex geological environments, where discontinuities vary significantly in scale, persistence, and spatial arrangement.

Geometric Feature–Driven Discontinuity Detection

The proposed method shifts focus from purely directional clustering to geometric feature analysis of rock mass point clouds. By examining spatial distribution variability and structural continuity, the framework captures both local and global geometric characteristics. An adaptive region-growing algorithm is integrated to detect independent discontinuities even under irregular rock mass geometries. This enables accurate segmentation across diverse rock shapes and sizes while minimizing sensitivity to noise.

Adaptive Hierarchical Clustering Based on Fisher Distribution

Recognizing that rock mass orientations typically follow a Fisher distribution, the study introduces a statistically grounded adaptive hierarchical clustering algorithm. Unlike conventional methods requiring preset cluster numbers, this approach automatically determines the optimal number of structural sets through statistical analysis of orientation dispersion. By eliminating manual intervention, the method enhances automation, objectivity, and reproducibility in discontinuity grouping.

Noise Resistance and Multi-Scale Feature Integration

A key strength of the framework lies in its integration of local geometric attributes (e.g., curvature, point density variations) with global structural trends. This multi-scale feature fusion reduces interference from measurement noise and incomplete data. The method demonstrates robustness in handling complex geological conditions, including intersecting discontinuities and variable persistence lengths, ensuring reliable detection across heterogeneous rock masses.

Experimental Validation and Engineering Significance

The proposed approach was validated using three real-world rock mass models and benchmarked against three mainstream directional clustering algorithms. Results indicate superior accuracy, improved detection of optimal discontinuity sets, and enhanced robustness under uneven orientation distributions. By offering a reliable and efficient tool for automated discontinuity detection and grouping, this method significantly strengthens geotechnical analysis, supporting safer underground excavation design and construction decision-making.

Global Civil Engineering Awards

🔗 Nominate now! 👉 https://civilengineeringawards.com/award-nomination/?ecategory=Awards&rcategory=Awardee

🌐 Visit: civilengineeringawards.com
📩 Contact: contact@civilengineeringawards.com

#EngineeringGeology

#3DModeling

#CivilEngineeringResearch

#GeometricAnalysis

#ConstructionSafety

#DigitalRockMass

#SmartGeotechnics

#InfrastructureDesign

PHYSICS-REGULARIZED DEEP LEARNING FOR INTER-FLOOR NOISE PREDICTION IN MULTI-FAMILY RESIDENTIAL BUILDINGS

 

Inter-floor noise in multi-family residential buildings represents a persistent structural-acoustic problem, particularly in densely populated urban environments. Excessive impact noise transmission not only contributes to occupant discomfort and health deterioration but also triggers legal disputes and regulatory challenges. Traditional acoustic prediction methods often rely on simplified empirical equations, while purely data-driven deep learning models lack physical interpretability and engineering reliability. Notably, prior studies have reported that 34–71% of predictions from conventional AI models violate fundamental acoustic laws. To address these shortcomings, this study proposes a physics-regularized multi-modal deep learning framework that integrates structural mechanics and acoustic principles into predictive modeling.

Structural-Acoustic Theoretical Foundations

The framework embeds governing physical principles of sound transmission in concrete floor systems directly into the learning process. Core mechanisms such as the Mass Law and the Coincidence Effect, derived from Kirchhoff–Love plate theory, are incorporated as differentiable constraints within the loss function. By enforcing these acoustic laws during optimization, the model ensures that predicted sound pressure levels remain consistent with structural vibration mechanics. This physics-informed strategy enhances interpretability and prevents non-physical prediction behavior commonly observed in black-box neural networks.

Multi-Modal Hybrid CNN–LSTM Architecture

A hybrid convolutional neural network (CNN) and long short-term memory (LSTM) architecture was developed to process multi-modal inputs. Structural parameters—including slab thickness, material properties, and span dimensions—are combined with acoustic spectrogram data representing frequency-domain characteristics of impact noise. The CNN component extracts spatial-frequency features from spectrograms, while the LSTM captures temporal dependencies and sequence behavior. This integration allows the model to simultaneously account for structural configuration and acoustic signal dynamics, improving predictive robustness across diverse building types.

Dataset and Cross-Validation Strategy

The proposed framework was validated using field measurements from 203 Korean multi-family residential buildings. To ensure generalization to unseen structural configurations, a Leave-One-Building-Out (LOBO) cross-validation scheme was adopted. This approach rigorously tests the model’s adaptability by training on all buildings except one and evaluating performance on the excluded structure. Such validation strengthens the reliability of results and demonstrates applicability across varying floor systems and construction conditions.

Prediction Accuracy and Physics Compliance

The physics-regularized model achieved prediction errors of 3.95 dB for light impact noise and 1.15 dB for heavy impact noise, representing approximately 10% improvement over baseline deep learning models. More importantly, physics violation rates were dramatically reduced from 71% to 1.8%, ensuring predictions adhere to structural mechanics and acoustic transmission principles. This substantial reduction confirms that embedding physical constraints effectively bridges the gap between predictive accuracy and engineering validity.

Regulatory Compliance and Engineering Implications

Beyond prediction accuracy, the model demonstrated 89.4% accuracy in forecasting compliance with the 58 dB Korean Building Code standard for inter-floor noise. This capability enables performance-based acoustic evaluation during early design stages, supporting proactive structural optimization and construction quality management. By uniting artificial intelligence with physically grounded constraints, the proposed framework establishes a new benchmark for reliable acoustic prediction in civil engineering, promoting safer, quieter, and regulation-compliant residential infrastructure.

🏗️ Civil Engineering Awards  

👉 Visit our Website: civilengineeringawards.com

#BuildingCode

#KirchhoffLoveTheory

#MassLaw

#CoincidenceEffect

#ConstructionQuality

#CivilEngineeringResearch

#SmartBuildings

#AIinConstruction

#ImpactNoise

#SustainableDesign

THE IMPACT OF GENERATIVE AI ON CIVIL ENGINEERING EDUCATION AND PROFESSIONAL PRACTICE

The rapid advancement of Generative AI (GenAI) is reshaping both educational and professional environments. From automated drafting to intelligent problem-solving support, AI-powered tools are enhancing productivity and enabling personalized learning experiences. However, while GenAI is widely recognized for improving efficiency, its influence on the learning process among civil engineering students and industry professionals remains relatively underexplored.

This study bridges that gap by examining how both groups perceive GenAI adoption, focusing on its impact on efficiency, problem-solving, motivation, teamwork, and future potential.

Research Methodology

The study employed a mixed-method approach:

• Quantitative analysis: Likert-scale survey responses

• Qualitative analysis: Thematic evaluation of open-ended feedback

Participants included civil engineering students and practicing professionals, allowing a cross-generational and cross-context comparison of perceptions.

Key Findings

1️⃣ Enhanced Efficiency

• 75% of participants viewed GenAI positively for improving efficiency.

• Respondents noted faster task completion, streamlined research, and improved access to technical explanations.

GenAI appears to act as a productivity amplifier in both academic and professional settings.

2️⃣ Improved Problem-Solving Skills

• 70% of participants identified problem-solving as the most improved skill.

• Students reported stronger perceived gains compared to professionals.

For students, GenAI provided:

• Step-by-step explanations

• Concept clarification

• Alternative solution strategies

Professionals, meanwhile, valued it for:

• Drafting assistance

• Code interpretation

• Technical summarization

3️⃣ Motivation and Learning Engagement

Students particularly reported increased motivation and confidence when using GenAI tools. Personalized feedback and immediate assistance reduced learning barriers and enhanced engagement.

This supports alignment with Sustainable Development Goal 4 (Quality Education) promoted by the United Nations.

4️⃣ Limited Impact on Teamwork

Interestingly, GenAI’s perceived effect on teamwork and collaboration was modest.

• Students reported minimal improvement in collaborative learning.

• Professionals observed slightly stronger collaboration advantages, especially in documentation and communication tasks.

This suggests that while GenAI enhances individual productivity, it does not automatically strengthen team dynamics.

Concerns and Ethical Considerations

Participants highlighted two primary concerns:

• Over-reliance on AI tools

• Accuracy and reliability of generated information

These concerns reinforce the importance of critical thinking and responsible AI use. GenAI should serve as a supplement, not a replacement, for foundational knowledge and professional judgment.

The study also connects to Sustainable Development Goal 8 (Decent Work and Economic Growth) by emphasizing workplace adaptability and future skill readiness.

Implications for Education and Industry

The findings underscore the importance of hybrid learning models, combining:

• Conventional teaching methods

• AI-assisted learning tools

Such an approach ensures:

✔ Retention of critical thinking skills
✔ Ethical AI integration
✔ Long-term professional competency
✔ Sustainable educational development

Educational institutions and engineering firms should develop policies encouraging guided, transparent, and responsible GenAI use.

Conclusion

Generative AI is widely perceived as a powerful efficiency and problem-solving enhancer in civil engineering education and practice. However, its integration must remain balanced, ethical, and critically supervised. By combining traditional methods with AI-assisted tools, educators and professionals can foster sustainable learning environments aligned with global development goals.

Future research should explore cross-context and longitudinal impacts to better understand how GenAI influences collaboration, skill development, and professional identity over time.

#SDG4
#SDG8
#WorkplaceInnovation
#HigherEducation
#SmartLearning

 

REAL-TIME ROCK MASS QUALITY PREDICTION USING ONLINE TRANSFER LEARNING IN TBM TUNNELLING

 

Accurate and real-time perception of rock mass quality during Tunnel Boring Machine (TBM) excavation is essential for ensuring operational safety, optimizing support measures, and minimizing construction risks. This challenge becomes even more critical in newly constructed tunnels, where limited prior geological investigations restrict the ability to promptly evaluate surrounding rock conditions. To address this issue, this study proposes a physical feature–shared online transfer learning framework that enables cross-engineering rock mass quality prediction using monitored data from previously completed tunnels.

Framework Overview

The proposed framework integrates three core components:

1️⃣ Online Data Stream Processor

A real-time data acquisition and processing module was developed to handle continuous TBM monitoring data. This processor performs:

• Real-time data collection

• Data segmentation

• Stream-based feature extraction

Importantly, it extracts physical shared features derived from the rock-cutting process, which are transferable across different tunnel projects due to their consistent physical meaning and accessibility.

2️⃣ Online Learning Module (LSTM-PSM)

An advanced classification module based on:

• Long Short-Term Memory (LSTM) networks

• Probabilistic State Machine (PSM)

was designed to estimate classification probability evolution in real time.

This hybrid LSTM-PSM model captures both temporal dependencies in TBM operational data and state transition characteristics of rock mass classifications, enabling dynamic and adaptive prediction.

3️⃣ Cross-Engineering Transfer Learning

The framework leverages completed tunnel data as a source domain and transfers knowledge to a new tunnel project (target domain), reducing reliance on extensive new-site training data.

Case Study Validation

The framework was validated using real-world engineering data:

• Source domain: Songhua River Tunnel (YS, 19.77 km)

• Target domain: Chaoer to Xiliao River Tunnel (YC, 2.27 km)

Field data from the longer, completed tunnel were used to train the transferable model, which was then applied to predict rock mass quality in the shorter, newly constructed tunnel.

Key Findings

The results demonstrate several significant advantages:

✔ Accurate Rock Mass Classification

The proposed online transfer learning framework accurately identifies:

• Rock mass quality grades

• State transition processes

in real-time excavation scenarios.

✔ Superior Physical Shared Features

Features derived from the rock-cutting process showed:

• Strong similarity across different tunnels

• High transferability

• Ease of acquisition during TBM operation

This makes them ideal candidates for cross-project intelligent modeling.

✔ Improved Model Performance

The online LSTM-PSM module outperformed state-of-the-art offline transfer learning methods in:

• Prediction accuracy

• Model reliability

• Adaptability to dynamic excavation conditions

Unlike offline methods, the proposed system continuously updates and adapts to evolving geological conditions.

Engineering Significance

This study introduces a novel approach for cross-engineering real-time rock mass quality perception. By combining physical feature extraction with online transfer learning, the framework:

• Enhances rock mass stability assessment

• Supports timely selection of support measures

• Improves tunnelling safety and efficiency

• Reduces dependence on extensive prior geological surveys

The approach represents a major step toward intelligent TBM excavation and digital tunnelling systems.

🏗️ Civil Engineering Awards  

👉 Visit our Website: civilengineeringawards.com

#InfrastructureInnovation

#MachineLearning

#RockMechanics

#CivilEngineering

#EngineeringGeology

#ConstructionTechnology