Applied Artificial Intelligence for Industrial Nondestructive Evaluation NDE4.0
1
Fraunhofer Institute for Nondestructive Testing IZFP, Campus E 3.1, 66123 Saarbruecken, Germany
2
School of Engineering, Saarland University of Applied Sciences, Goebenstrasse 40, 66117 Saarbrücken, Germany
3
Laboratory of Vibration and Acoustics, Mechanical Engineering and Design Department, National Institute of Applied Sciences of Lyon, INSA-Lyon, UR677, CEDEX, 69621 Villeurbanne, France
*
Author to whom correspondence should be addressed.
Appl. Sci. 2025, 15(9), 4968; https://doi.org/10.3390/app15094968
Submission received: 18 April 2025
/
Accepted: 26 April 2025
/
Published: 30 April 2025
(This article belongs to the Section Applied Industrial Technologies)
1. Introduction
Imagine a world where a single undetected crack in a turbine blade or a hidden flaw in a 3D-printed component could halt production or worse. This is the stakes-driven reality that nondestructive evaluation (NDE) confronts, now revolutionized by artificial intelligence (AI) and deep learning (DL) in the era of NDE 4.0. By fusing advanced sensors, robotics, and intelligent data analysis, NDE 4.0 empowers industries, from aerospace to energy, to detect defects with unprecedented precision. This Special Issue “Applied Artificial Intelligence for Industrial Nondestructive Evaluation NDE4.0” addresses the following challenges: noisy data drowning out signals, complex defects that are difficult to interpret, and the struggle of AI to scale or earn trust, with 11 pioneering papers, illuminating solutions and igniting curiosity for what lies ahead.
2. An Overview of the Published Articles
This collection showcases AI’s transformative potential across NDE applications, from factory floors to infrastructure maintenance. These 11 contributions, diverse yet interconnected, address critical gaps with ingenuity and practical relevance.
Lang et al. (contribution 1) harnessed deep learning to monitor additive co-extrusion processes in real time using laser profiler data. Their work strengthened thermoplastic components—vital for industries such as the automotive—by detecting defects as they form in advanced manufacturing settings.
Bai et al. (contribution 2) blended temporal convolutional networks with bidirectional LSTM to predict equipment faults, mastering the complexity of time-series data and addressing limitations in traditional fault prediction methods.
Wang and Xie (contribution 3) pivoted to acoustic signals for rolling bearing diagnostics in noisy environments, leveraging microphones’ affordability and sensitivity over vibration sensors to tackle signal interference.
Ko and Lee (contribution 4) introduced a multi-patch time series transformer to detect bearing faults under varying noise, mimicking the chaotic conditions of industrial settings with novel noise techniques.
Jin et al. (contribution 5) used multiple attention mechanisms to segment defects in drainage pipeline sonar images, safeguarding aging infrastructure from collapse by targeting complex defect types.
Li and Chen (contribution 6) deployed machine learning, optimized by particle swarm optimization, to predict lithium-ion battery health—a linchpin for sustainable energy systems—using LSTM, CNN, and SVR models.
Wang et al. (contribution 7) elevated steel production with an improved YOLOv8 model, tackling the complex reality of surface defects amid noise and complex backgrounds.
Val et al. (contribution 8) turned to AI to analyze wear in milling insert tools, optimizing lifespans and slashing costs in modern machining by examining flank wear patterns.
Baumeyer et al. (contribution 9) enhanced magnetic particle inspection with AI assistance, improving defect detection under UV-A lighting—a boon for quality control in harsh industrial environments.
Wei et al. (contribution 10, Appl. Sci. 2023, 13(24), 13093) provided a pulsed thermography dataset for automated defect depth estimation in PVC, addressing the challenge of subsurface defect characterization.
Wei et al. (contribution 11, Appl. Sci. 2023, 13(5), 2901) offered another dataset for training DL models in pulsed thermography, easing the interpretive burden and fueling research with high-quality data.
Together, these papers bridge noisy data, defect complexity, and scalability gaps, blending AI with human expertise to redefine industrial reliability.
3. Conclusions
The journey of NDE 4.0 is just beginning, and this Special Issue paves the way toward new horizons. How can we ensure AI earns the trust of engineers and regulators in life-or-death applications? Developing explainable models and universal validation metrics is a pressing next step. As sensor networks flood us with data, scaling NDE through cloud, edge, and fog computing beckons as an urgent challenge—and opportunity. The partnership between human inspectors and AI assistants remains uncharted; optimizing this relationship could unlock new efficiencies. With industries racing toward composites and 3D-printed parts, adapting these methods to novel materials is also worthwhile.
We invite researchers, engineers, and industry pioneers to address these questions. Imagine an NDE 4.0 that does not just detect flaws but predicts and prevents them—saving time, costs, and lives. This collection is your springboard; dive in, explore, and shape the future of this vibrant field.
Author Contributions
A.O.: Writing—original draft, Writing—review and editing. V.K.: Review and editing. All authors have read and agreed to the published version of the manuscript.
Conflicts of Interest
The authors declare no conflicts of interest.
List of Contributions
- Lang, V.; Herrmann, C.; Fuchs, M.; Ihlenfeldt, S. Deep Learning Utilization for In-Line Monitoring of an Additive Co-Extrusion Process Based on Evaluation of Laser Profiler Data. Appl. Sci. 2025, 15, 1727. https://doi.org/10.3390/app15041727.
- Bai, J.; Zhu, W.; Liu, S.; Ye, C.; Zheng, P.; Wang, X. A Temporal Convolutional Network–Bidirectional Long Short-Term Memory (TCN-BiLSTM) Prediction Model for Temporal Faults in Industrial Equipment. Appl. Sci. 2025, 15, 1702. https://doi.org/10.3390/app15041702.
- Wang, H.; Xie, J. Fault Diagnosis of Rolling Bearings Based on Acoustic Signals in Strong Noise Environments. Appl. Sci. 2025, 15, 1389. https://doi.org/10.3390/app15031389.
- Ko, S.; Lee, S. Multi-Patch Time Series Transformer for Robust Bearing Fault Detection with Varying Noise. Appl. Sci. 2025, 15, 1257. https://doi.org/10.3390/app15031257.
- Jin, Q.; Han, Q.; Qian, J.; Sun, L.; Ge, K.; Xia, J. Drainage Pipeline Multi-Defect Segmentation Assisted by Multiple Attention for Sonar Images. Appl. Sci. 2025, 15, 597. https://doi.org/10.3390/app15020597.
- Li, K.; Chen, X. Machine Learning-Based Lithium Battery State of Health Prediction Research. Appl. Sci. 2025, 15, 516. https://doi.org/10.3390/app15020516.
- Wang, J.; Chen, T.; Xu, X.; Zhao, L.; Yuan, D.; Du, Y.; Guo, X.; Chen, N. An Improved YOLOv8 Model for Strip Steel Surface Defect Detection. Appl. Sci. 2025, 15, 52. https://doi.org/10.3390/app15010052.
- Val, S.; Lambán, M.; Lucia, J.; Royo, J. Analysis and Prediction of Wear in Interchangeable Milling Insert Tools Using Artificial Intelligence Techniques. Appl. Sci. 2024, 14, 11840. https://doi.org/10.3390/app142411840.
- Baumeyer, J.; Chatoux, H.; Pelletier, A.; Marquié, P. Industrial Application of AI-Based Assistive Magnetic Particle Inspection. Appl. Sci. 2024, 14, 1499. https://doi.org/10.3390/app14041499.
- Wei, Z.; Osman, A.; Valeske, B.; Maldague, X. A Dataset of Pulsed Thermography for Automated Defect Depth Estimation. Appl. Sci. 2023, 13, 13093. https://doi.org/10.3390/app132413093.
- Wei, Z.; Osman, A.; Valeske, B.; Maldague, X. Pulsed Thermography Dataset for Training Deep Learning Models. Appl. Sci. 2023, 13, 2901. https://doi.org/10.3390/app13052901.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
MDPI and ACS Style
Osman, A.; Kaftandjian, V. Applied Artificial Intelligence for Industrial Nondestructive Evaluation NDE4.0. Appl. Sci. 2025, 15, 4968. https://doi.org/10.3390/app15094968
AMA Style
Osman A, Kaftandjian V. Applied Artificial Intelligence for Industrial Nondestructive Evaluation NDE4.0. Applied Sciences. 2025; 15(9):4968. https://doi.org/10.3390/app15094968
Chicago/Turabian StyleOsman, Ahmad, and Valerie Kaftandjian. 2025. "Applied Artificial Intelligence for Industrial Nondestructive Evaluation NDE4.0" Applied Sciences 15, no. 9: 4968. https://doi.org/10.3390/app15094968
APA StyleOsman, A., & Kaftandjian, V. (2025). Applied Artificial Intelligence for Industrial Nondestructive Evaluation NDE4.0. Applied Sciences, 15(9), 4968. https://doi.org/10.3390/app15094968
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.
