A framework based on deep learning and the intelligent sensors for pavement assessment condition

Wael A. Altabey

doi:10.59400/sv3458

A framework based on deep learning and the intelligent sensors for pavement assessment condition

Wael A. Altabey
Department of Mechanical Engineering, Faculty of Engineering, Alexandria University, Alexandria 21544, Egypt

Article ID: 3458

Keywords: Structural Health Monitoring (SHM); long-term pavement performance; pavement sensors; Internet of Things (IoT); deep learning, convolutional neural network

Abstract

Long-term pavement performance is a key topic in highway engineering. By diving deep into research on pavement systems, we can bring together past, fragmented knowledge and experiences into a solid, comprehensive engineering theory. This essentially helps guide practical work like pavement design, construction, maintenance, and management. In this research, we look at using a mentoring system for automatic monitoring of pavement performance. By placing various sensors in different positions like the road surface, base, and slopes, a sensor network powered by Internet of Things technology is created. This setup allows for accurate and ongoing observation of factors like weather, physical condition, mechanical responses, and structural changes. Given the large volume of data and the need for real-time analysis, a data from sensors measuring temperature, humidity, pressure, asphalt strain, and displacement are used to train a deep learning model based on a Convolutional Neural Network (CNN) algorithm. This model helps predict multi-point displacement in the pavement, which allows us to detect issues like pavement damage. Impressively, the CNN model achieved accuracy, regression rates, and F-score of 93.51%, 91.63%, and 90.64% respectively. To improve the experimental section of a deep learning study, we compared the performance of the proposed model against several established or simpler algorithms (baselines) in the literature such as K-Nearest Neighbors (K-NN), eXtreme Gradient Boosting (XGBoost), and support vector machine (SVM). This contextualizes the model's efficacy and demonstrates its advantage over existing methods. This study showcases how different sensors can support deep learning algorithms in the assessment of pavement performance over the long term.

Published

2025-11-04

How to Cite

Altabey, W. A. (2025). A framework based on deep learning and the intelligent sensors for pavement assessment condition. Sound & Vibration, 59(6). https://doi.org/10.59400/sv3458

Download Citation

Issue

Vol. 59 No. 6 (2025)

Section

Article

This work is licensed under a Creative Commons Attribution 4.0 International License.

References

[1]Shtayat A, Moridpour S, Best B. Using e-bikes and private cars in dynamic road pavement monitoring. International Journal of Transportation Science and Technology. 2022; 11(1): 132–143. doi: 10.1016/j.ijtst.2021.03.004

[2]Justo-Silva R, Ferreira A, Flintsch G. Review on Machine Learning Techniques for Developing Pavement Performance Prediction Models. Sustainability. 2021; 13(9): 5248. doi: 10.3390/su13095248

[3]Altabey WA, Kouritem SallamA, Al-Moghazy MA. Apply frequency response function schemes for damage detection in composite nanoscale-pipes under transient conditions. Nano-Structures & Nano-Objects. 2024; 39: 101259. doi: 10.1016/j.nanoso.2024.101259

[4]Altabey WA, Noori M. A Mathematical Modeling of BFRP Laminated Composite Double-Chamber Mufflers Based Acoustic Transmission Loss Optimization. In: Proceedings of the International Conference of Steel and Composite for Engineering Structures; 20–21 November 2023; Lecce, Italy. pp. 54–65. doi: 10.1007/978-3-031-57224-1_6

[5]Altabey WA. An Artificial Intelligence-Based Scheme for Structural Health Monitoring in CFRE Laminated Composite Plates under Spectrum Fatigue Loading. Structural Durability & Health Monitoring. 2025; 19(5): 1145–1165. doi: 10.32604/sdhm.2025.068922

[6]Haja Mohideen AJ, Rosli MF, Mohamad Hanif NHH, et al. Pavement condition analysis via vehicle mounted accelerometer data. IIUM Engineering Journal. 2020; 21(1): 73–84. doi: 10.31436/iiumej.v21i1.1223

[7]Farhadmanesh M, Cross C, Mashhadi AH, et al. Highway Asset and Pavement Condition Management using Mobile Photogrammetry. Transportation Research Record: Journal of the Transportation Research Board. 2021; 2675(9): 296–307. doi: 10.1177/03611981211001855

[8]Arezoumand S, Mahmoudzadeh A, Golroo A, et al. Automatic pavement rutting measurement by fusing a high speed-shot camera and a linear laser. Construction and Building Materials. 2021; 283: 122668. doi: 10.1016/j.conbuildmat.2021.122668

[9]Jahanshahi MR, Jazizadeh F, Masri SF, et al. Unsupervised Approach for Autonomous Pavement-Defect Detection and Quantification Using an Inexpensive Depth Sensor. Journal of Computing in Civil Engineering. 2013; 27(6): 743–754. doi: 10.1061/(ASCE)CP.1943-5487.0000245

[10]Buza E, Akagic A, Omanovic S, et al. Unsupervised method for detection of high severity distresses on asphalt pavements, In: Proceedings of the 2017 IEEE 14th International Scientific Conference on Informatics; 14–16 November 2017; Poprad, Slovakia. pp. 45–50. doi: 10.1109/INFORMATICS.2017.8327220

[11]Lekshmipathy J, Samuel NM, Velayudhan S. Vibration vs. vision: best approach for automated pavement distress detection. International Journal of Pavement Research and Technology. 2020; 13(4): 402–410. doi: 10.1007/s42947-020-0302-y

[12]Zhang C, Nateghinia E, Miranda-Moreno LF, et al. Pavement distress detection using convolutional neural network (CNN): A case study in Montreal, Canada. International Journal of Transportation Science and Technology. 2022; 11(2): 298–309. doi: 10.1016/j.ijtst.2021.04.008

[13]Koch C, Georgieva K, Kasireddy V, et al. A review on computer vision based defect detection and condition assessment of concrete and asphalt civil infrastructure. Advanced Engineering Informatics. 2015; 29(2): 196–210. doi: 10.1016/j.aei.2015.01.008

[14]Hadjidemetriou GM, Vela PA, Christodoulou SE. Automated Pavement Patch Detection and Quantification Using Support Vector Machines. Journal of Computing in Civil Engineering. 2018; 32(1): 04017073. doi: 10.1061/(ASCE)CP.1943-5487.0000724

[15]Cano-Ortiz S, Pascual-Muñoz P, Castro-Fresno D. Machine learning algorithms for monitoring pavement performance. Automation in Construction. 2022; 139: 104309. doi: 10.1016/j.autcon.2022.104309

[16]Di Graziano A, Marchetta V, Cafiso S. Structural health monitoring of asphalt pavements using smart sensor networks: A comprehensive review. Journal of Traffic and Transportation Engineering (English Edition). 2020; 7(5): 639–651. doi: 10.1016/j.jtte.2020.08.001

[17]Zhang C, Yan Q, Liao X, et al. Deep hybrid neural network-aided electromechanical impedance method for automated damage detection of lining concrete under freeze-thaw cycling. Structural Health Monitoring. 2025; 24(3): 1725–1751. doi: 10.1177/14759217241259955

[18]Ai D, Mo F, Cheng J, et al. Deep learning of electromechanical impedance for concrete structural damage identification using 1-D convolutional neural networks. Construction and Building Materials. 2023; 385: 131423. doi: 10.1016/j.conbuildmat.2023.131423

[19]Sapidis GM, Naoum MC, Papadopoulos NA, et al. A Novel Approach to Monitoring the Performance of Carbon-Fiber-Reinforced Polymer Retrofitting in Reinforced Concrete Beam–Column Joints. Applied Sciences. 2024; 14(20): 9173. doi: 10.3390/app14209173

[20]Ai D, Mo F, Han Y, et al. Automated identification of compressive stress and damage in concrete specimen using convolutional neural network learned electromechanical admittance. Engineering Structures. 2022; 259: 114176. doi: 10.1016/j.engstruct.2022.114176

[21]Sapidis GM, Kansizoglou I, Naoum MC, et al. A Deep Learning Approach for Autonomous Compression Damage Identification in Fiber-Reinforced Concrete Using Piezoelectric Lead Zirconate Titanate Transducers. Sensors. 2024; 24(2): 386. doi: 10.3390/s24020386

[22]Ai D, Mo F. Prediction of unknown compressive stress/damage in concrete structure using 3D convolutional neural network-based deep learning of raw electromechanical admittance signals. Journal of Intelligent Material Systems and Structures. 2025; 36(1): 29–52. doi: 10.1177/1045389X241291214

[23]Lin C-F, Zhu J-D. Hilbert–Huang transformation-based time-frequency analysis methods in biomedical signal applications. Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine. 2012; 226(3): 208–216. doi: 10.1177/0954411911434246

[24]Zhu X, Cao M, Ostachowicz W, et al. Damage Identification in Bridges by Processing Dynamic Responses to Moving Loads: Features and Evaluation. Sensors. 2019; 19(3): 463. doi: 10.3390/s19030463

[25]Yazdi A, Dehnad MH. A case study on predicting asphalt pavement distress using advanced machine learning techniques and road surface inspection data. Case Studies in Construction Materials. 2025; 23: e05291. doi: 10.1016/j.cscm.2025.e05291

This paper introduces a universal framework for understanding the vibration responses of systems subjected to harmonic excitation. By examining a simplified cylinder-spring-damper model, the study refurbishes traditional scaling methods for the excitation frequency ratio and critical damping ratio. The findings indicate that in damped systems, the maximum amplitude of vibration does not align with the natural frequency. This observation leads to the introduction of a new scaling method for reduced frequency. This new approach aligns resonance peaks at the new reduced velocity of 1.0 across different damping ratios, providing a consistent characterization of vibration behavior. A new critical damping ratio of 0.707 is identified for an excited system as opposed to the traditional damping ratio of 1.0 for an unexcited system. Key properties such as maximum amplitude, phase lag, bandwidth, and quality factor are analyzed, demonstrating that the proposed reduced frequency and critical damping ratio effectively capture the dynamics of both damped and undamped excited systems. The findings offer significant insights for practical applications in engineering and various scientific fields.

This study makes significant contributions to the field of ultrasonic testing (UT) by offering a novel approach to the identification of artificially introduced defects within Japanese cedar wood (Cryptomeria japonica). The findings are of particular relevance for the heritage conservation and construction sectors, where non-invasive defect detection is paramount. The study establishes a robust framework for assessing the structural integrity of timber by correlating ultrasonic wave velocity reductions with defect size and distribution. Big-sized defects led to more substantial decreases in wave velocity. The study establishes a robust framework for assessing the structural integrity of historical timber by correlating ultrasonic wave velocity reductions with defect size and distribution. This framework has the potential to be applicable to diverse wood species and defect types.

The control of vehicle interior noise has become a critical metric for assessing noise, vibration, and harshness (NVH) in vehicles. During the initial phases of vehicle development, accurately predicting the impact of road noise on interior noise is essential for reducing noise levels and expediting the product development cycle. In recent years, data-driven methods based on machine learning have gained significant attention due to their robust capability in navigating complex data mapping relationships. Notably, surrogate models have demonstrated exceptional performance in this domain. Numerous researchers have integrated diverse intelligent algorithms into the study of vehicle noise, leveraging advantages such as the elimination of precise modeling requirements, extensive solution space exploration, continuous learning from data, and robust algorithmic versatility. However, in NVH engineering applications, data-driven models face inherent limitations, particularly in interpretability and stability. To address these issues, this paper introduces an improved Long Short-Term Memory (LSTM) network that combines knowledge and data. Inspired by the physical information neural network concept, this approach incorporates values calculated through empirical formulas into the neural network as constraints. Comparative assessments with traditional LSTM networks highlight the advantages of this deep learning model. By integrating empirical formulas constraints, the model not only enhances interpretability but also achieves robust generalization with fewer data samples. The proposed method is validated on a specific vehicle model, showing significant improvements in prediction accuracy and efficiency.