Open research questions in Infrastructure Maintenance and Monitoring

What the literature leaves open

• Future research could explore integrating three-dimensional geometric information, such as LiDAR point clouds, to overcome the limitations of two-dimensional vision and improve reliability in scenarios where visual features are insufficient.

  An Automatic Detection Method for Coal–Rock Interface Based on YOLOv8 and Cubic Spline Interpolation · 2026 · DOI

• Future work will investigate multi-view building imagery, where multiple photographs of the same structure from different viewpoints are jointly analyzed.

  AI-Based Image and Data Analysis for Automated Assessment of Residential Damage in Seismic Regions · 2026 · DOI

• Future work will focus on the following three directions: (1) Extending the MDD framework to multi-modal crack detection tasks, such as thermal infrared and depth images, to leverage multi-source information complementarity and overcome the perceptual limitations of single modalities in extreme lighting or camouflaged scenarios.

  Enhancing crack detection via memory-aware dynamic knowledge distillation · 2026 · DOI

• Future research will investigate more efficient constraint-handling mechanisms that guide the search toward feasibility by explicitly prioritizing feasibility and minimizing constraint violation, following feasibility-rule-based principles in constrained evolutionary optimization and robustness-oriented strategies that are effective for stochastic evolutionary optimization in discrete/binary decision spaces.

  Spatiotemporal Particle Swarm Optimization for Future Cost Allocation in Large-Scale Transportation Infrastructure Maintenance · 2026 · DOI

• Although the proposed method achieves improvements on the reported RDD2022 split, several limitations remain.

  FasterGeo-YOLO: Multi-Scale Geometric Adaptive Feature Enhancement for Lightweight Road Damage Detection · 2026 · DOI

• The limitations of this study are as follows: (1) Tests were conducted on the same hardware platform, which precluded a comparative analysis of applications deployed on different platforms. (2) The study lacked validation using a diverse and extensive dataset. Additionally, there was insufficient research on the impact of data quality on the results.

  Comparative Analysis of YOLO Series Algorithms for UAV-Based Highway Distress Inspection: Performance and Application Insights · 2025 · DOI

• Deep learning U-Net ++ algorithm: Data dependent: requires high diversity and representativeness of the training dataset; insufficient data in complex conditions can lead to overfitting or poor generalization.

  A Bridge Crack Segmentation Algorithm Based on Fuzzy C-Means Clustering and Feature Fusion · 2025 · DOI

• Future research should consider applying this method to other object detection algorithms to further assess its generalizability. Future work will focus on further refining the model and improving detection performance to promote the widespread adoption of this method in the industrial sector. This method offers a fresh perspective on addressing the challenge of insufficient data in industrial inspection tasks.

  Enhancing Bolt Object Detection via AIGC-Driven Data Augmentation for Automated Construction Inspection · 2025 · DOI

• The findings demonstrate that while baseline CNNs can achieve moderate classification accuracy, their ability to generalize across diverse surface conditions and capture subtle structural anomalies remains limited. The use of class-weighted loss significantly improved minority class detection, reaffirming its importance in imbalanced real-world datasets.

  Deployable Deep Learning Models for Crack Detection: Efficiency, Interpretability, and Severity Estimation · 2025 · DOI

• Future research could explore the integration of advanced machine learning and deep learning models, such as CNNs or graph neural networks (GNNs), which are well suited for handling high-dimensional data and may enhance the detection of complex crack patterns, particularly transverse cracks.

  LiDAR-Based Road Cracking Detection: Machine Learning Comparison, Intensity Normalization, and Open-Source WebGIS for Infrastructure Maintenance · 2025 · DOI

• This suggests that future research should explore hybrid approaches combining unsupervised clustering and supervised learning techniques, as the integration of both methods may enhance crack detection by leveraging the strengths of each: unsupervised clustering for identifying patterns in complex data and supervised learning for improving classification accuracy across different crack types.

  LiDAR-Based Road Cracking Detection: Machine Learning Comparison, Intensity Normalization, and Open-Source WebGIS for Infrastructure Maintenance · 2025 · DOI

• Future studies should focus on improving algorithms, optimizing computational processes, and collecting diversified datasets, ensuring that these technologies can be effectively applied in a variety of real-world scenarios and conditions.

  PRISMA Review: Drones and AI in Inventory Creation of Signage · 2025 · DOI

• The goal of this evaluation was not to establish a comprehensive benchmark—which falls beyond the scope of this study—but, rather, to support the positioning of our design among existing approaches. This highlights a key limitation: while effective for segmentation, non-learnable enhancements may conflict with detection-level optimization due to their rigid, non-adaptive nature.

  PowerLine-MTYOLO: A Multitask YOLO Model for Simultaneous Cable Segmentation and Broken Strand Detection · 2025 · DOI

• The lack of a widely accepted standard or reference model in the field of road infrastructure management. The need for digital twin standards that are applicable across business domains to facilitate interoperability. The need for digital twin interoperability standards that include reference models, architectures, and protocols.

  Digital Twin Framework for Road Infrastructure Management · 2025 · DOI

• The need for further research and optimization in the area of road debris identification.

  Urban Road Anomaly Monitoring Using Vision–Language Models for Enhanced Safety Management · 2025 · DOI

• The need for enhanced road safety measures to mitigate the risks associated with urban road anomalies. The lack of real-time monitoring and early warning systems for urban road anomalies.

  Urban Road Anomaly Monitoring Using Vision–Language Models for Enhanced Safety Management · 2025 · DOI

• Conventional road damage assessments rely heavily on manual inspection, which is labor-intensive and time-consuming. Standard training strategies are insufficient to address the severe class imbalance inherent in real-world datasets.

  Deep Learning-Based Detection and Assessment of Road Damage Caused by Disaster with Satellite Imagery · 2025 · DOI

• Future research could focus on developing advanced cybersecurity measures to protect against threats and vulnerabilities. Transfer learning and domain adaptation Study the application of transfer learning and domain adaptation techniques to improve the performance of image classification algorithms when applied to concrete fault detection in bridges, especially in cases where labeled data is scarce.

  Improving the Concrete Crack Detection Process via a Hybrid Visual Transformer Algorithm · 2024 · DOI

• Limited high-quality datasets for AI/ML applications in civil engineering. Absence of AI provisions in design codes. Integration barriers with IoT-based infrastructure.

  The Role of Artificial Intelligence and Machine Learning in Advancing Civil Engineering: A Comprehensive Review · 2025 · DOI

• The Adam learning algorithm is employed, and it achieves an accuracy of 94.17 ± 1.67% on the test set with a validation loss of 0.1976.

  Tiny-Machine-Learning-Based Supply Canal Surface Condition Monitoring · 2024 · DOI

• Federated learning for data privacy, - Transfer learning for data-scarce regions, - Digital twins, - Adherence to FAIR data…

  The Role of Artificial Intelligence and Machine Learning in Advancing Civil Engineering: A Comprehensive Review · 2025 · DOI

• PDPs have a limitation in the spatial dimension, which can only respond to two features at most - PDPs need to satisfy the assumption that the features are independent of each…

  An Explainable XGBoost Model for International Roughness Index Prediction and Key Factor Identification · 2025 · DOI

• Future research must focus on methods that directly address class imbalance to improve detection recall - Developing strategies to address severe class imbalance…

  Deep Learning-Based Detection and Assessment of Road Damage Caused by Disaster with Satellite Imagery · 2025 · DOI

• Investigating the effect of different image categories on the performance of GoogleNet and SegNet - Examining the impact of iteration count, image quantity, and image category on the performance of GoogleNet and SegNet - Exploring the application of deep learning-based crack detection in other…

  Deep Learning-Based Crack Detection on Cultural Heritage Surfaces · 2025 · DOI

• The study identifies a gap in the current research trends, key contributing countries, and emerging topics in AI-integrated SHM. The study also identifies a gap in the ethical considerations and societal impacts of AI in SHM.

  AI in Structural Health Monitoring for Infrastructure Maintenance and Safety · 2024 · DOI

Most-cited papers in Infrastructure Maintenance and Monitoring

• Feature Pyramid and Hierarchical Boosting Network for Pavement Crack Detection · IEEE Transactions on Intelligent Transportation Systems · 2020 · 1,188 citations
• Toward safer highways, application of XGBoost and SHAP for real-time accident detection and feature analysis · Accident Analysis & Prevention · 2019 · 1,004 citations
• Machine learning for structural engineering: A state-of-the-art review · Structures · 2022 · 677 citations
• Application of deep learning algorithms in geotechnical engineering: a short critical review · Artificial Intelligence Review · 2021 · 466 citations
• Interpretable XGBoost-SHAP Machine-Learning Model for Shear Strength Prediction of Squat RC Walls · Journal of Structural Engineering · 2021 · 423 citations
• A critical review and comparative study on image segmentation-based techniques for pavement crack detection · Construction and Building Materials · 2022 · 409 citations
• Automatic crack classification and segmentation on masonry surfaces using convolutional neural networks and transfer learning · Automation in Construction · 2021 · 408 citations
• Hybrid pixel-level concrete crack segmentation and quantification across complex backgrounds using deep learning · Automation in Construction · 2020 · 408 citations
• Machine Learning for Crack Detection: Review and Model Performance Comparison · Journal of Computing in Civil Engineering · 2020 · 407 citations
• Applications of unmanned aerial vehicle (UAV) in road safety, traffic and highway infrastructure management: Recent advances and challenges · Transportation Research Part A Policy and Practice · 2020 · 393 citations

Most recent work

• UAV-based quantitative crack measurement for bridges integrating four-point laser metric calibration and mamba segmentation · Automation in Construction · 2026
• A novel hybrid algorithm for damage detection in bridge foundations under complex underwater environments using ROV capture pictures · Engineering Structures · 2026
• A visual attention-guided approach for concrete crack detection in complex environments · Engineering Applications of Artificial Intelligence · 2026
• Toward intelligent pavement maintenance: A transferable deep learning framework for cross-domain crack segmentation and UAV-based field inspection · Advanced Engineering Informatics · 2026
• BCLIP-ADer: A Bayesian Prompt Contrastive Language-Image Pretraining Method for Catenary Component Anomaly Detection in Electrified Railways · IEEE Transactions on Transportation Electrification · 2026
• BrIMs-Based 3D Semantic Segmentation of Bridge Components Leveraging Multisensor Fusion · Journal of Computing in Civil Engineering · 2026
• Integrating segmentation and vision-language model for automated and interpretable building damage assessment from satellite imagery · Advanced Engineering Informatics · 2026
• Multimodal large language model-driven framework for road crack assessment · Automation in Construction · 2026
• From instance segmentation to physical quantification: High-resolution UAV-based dataset for façade defect assessment · Automation in Construction · 2026
• A fusion framework of improved YOLOv8 and lightweight DeepLabv3 + for concrete bridge surface defect identification · Engineering Fracture Mechanics · 2026

Related topics in Engineering