AI-Based Defect Identification in FDM-Printed Biodegradable Polymer Composites Through Multimodal Characterization
DOI:
https://doi.org/10.12974/2311-8717.2025.13.15Keywords:
FDM, Biodegradable polymer composites, Defect identification, Multimodal characterization, Convolutional neural network, Surface analysis, Additive manufacturingAbstract
The growing demand for biodegradable polymer composites in sustainable manufacturing requires robust quality-assessment frameworks that ensure structural reliability and functional performance. However, FDM-based additive manufacturing of such materials often introduces processing-induced defects that compromise mechanical integrity. Conventional visual inspection remains subjective and limited, creating a need for advanced, automated defect-identification strategies. This study addresses this challenge by integrating artificial intelligence with multimodal characterization to establish a reliable defect-detection pipeline for FDM-printed biodegradable polymer composites. Biodegradable PLA-based composites reinforced with microscale and nanoscale fillers were fabricated under controlled FDM conditions, followed by systematic defect mapping through optical imaging, SEM, and surface profilometry. A convolutional neural-network classifier was trained using 2,500 labelled images, incorporating multimodal inputs to identify four major defects: voids, layer gaps, surface roughness irregularities, and under-extrusion patterns. The optimized AI model achieved an overall classification accuracy of 96.4%, precision of 94.8%, recall of 95.3%, and an F1-score of 95.0%, outperforming traditional threshold-based and handcrafted-feature methods. Multimodal correlation analysis further revealed that defects predicted with high probability aligned strongly with SEM-verified structural anomalies (R² = 0.93) and surface-roughness deviations (up to 18% variation). These results demonstrate that AI-assisted evaluation offers a reliable, scalable, and non-destructive pathway to improve defect quantification in biodegradable polymer composites. The proposed framework enhances process monitoring, reduces inspection subjectivity, and provides new insights into structure–processing–defect interrelationships in FDM-printed sustainable composites.
References
Vidakis, N., Michailidis, N., Kalderis, D., Argyros, A., Gkagkanatsiou, K., Spyridaki, M., ... & Petousis, M. (2025). Biodegradable polyhydroxyalkanoate (PHA) composites with biochar ratios optimized for the additive manufacturing method of material extrusion: engineering, rheological, and morphological insights. Materials Advances, 6(18), 6427-6444. https://doi.org/10.1039/D5MA00266D
Farooq, E., Osama, S. M., Abbas, S. H., & Aqeel, M. (2025). Biodegradable Polymers for Process Intensification in Chemical Engineering: Challenges and Innovations. Mechanics Exploration and Material Innovation, 2(1), 1-13. https://doi.org/10.21776/ub.memi.2025.002.01.1
Mantri, R. K., Gatade, A. T., Chakule, R. R., Aswar, S. J., Kharche, Y. A., Khandare, N., & Baviskar, D. (2025). The Evolution of Additive Manufacturing in Sustainable Biomaterial Production. Journal of Mines, Metals and Fuels, 73(8), 2427-2437. https://doi.org/10.18311/jmmf/2025/49404
Erokhin, K., Naumov, S., & Ananikov, V. (2023). Defects in 3D printing and strategies to enhance quality of FFF additive manufacturing. A Review. https://doi.org/10.26434/chemrxiv-2023-lw1ns
Massaroni, C., Saroli, V., Aloqalaa, Z., Presti, D. L., & Schena, E. (2025). Extrusion‐Based Fused Deposition Modeling for Printing Sensors and Electrodes: Materials, Process Parameters, and Applications. SmartMat, 6(4), e70027. https://doi.org/10.1002/smm2.70027
Malashin, I., Masich, I., Tynchenko, V., Gantimurov, A., Nelyub, V., Borodulin, A., ... & Galinovsky, A. (2024). Machine learning in 3D and 4D printing of polymer composites: a review. Polymers, 16(22), 3125. https://doi.org/10.3390/polym16223125
Salehiyan, R., & Soleymani Eil Bakhtiari, S. (2024). A review on rheological approaches as a perfect tool to monitor thermal degradation of biodegradable polymers. Korea-Australia Rheology Journal, 36(4), 295-317. https://doi.org/10.1007/s13367-024-00111-3
Niño, J. P. C., Hernandez, H. M. J., Florido, H. A. F., López, L. M., Sánchez, M. D. R. S., & Duque, J. F. S. (2023). Rheology Properties of Biodegradable Polymers. In Biodegradable Polymers (pp. 80-98). CRC Press. https://doi.org/10.1201/9781003230533-6
Li, Y. D., Zeng, J. B., Li, W. D., Yang, K. K., Wang, X. L., & Wang, Y. Z. (2009). Rheology, crystallization, and biodegradability of blends based on soy protein and chemically modified poly (butylene succinate). Industrial & engineering chemistry research, 48(10), 4817-4825. https://doi.org/10.1021/ie801718f
Bonato, N., Zanini, F., & Carmignato, S. (2025). Enhanced tomographic porosity measurements in laser powder bed fusion metal parts using a novel reference object. Journal of Manufacturing Processes, 155, 1049-1061. https://doi.org/10.1016/j.jmapro.2025.10.076
Akbas, O., Gaikwad, A., Reck, L., Ehlert, N., Jahn, A. M., Hermsdorf, J., ... & Greuling, A. (2025). Effects of sandblasting and acid etching on the surface properties of additively manufactured and machined titanium and their consequences for osteoblast adhesion under different storage conditions. Frontiers in Bioengineering and Biotechnology, 13, 1640122. https://doi.org/10.3389/fbioe.2025.1640122
Taieb, K. (2025). Parametric study of precision abrasive aachining arocesses for ultra-precise finishing (Doctoral dissertation).
Karuppusamy, M., Thirumalaisamy, R., Palanisamy, S., Nagamalai, S., Massoud, E. E. S., & Ayrilmis, N. (2025). A review of machine learning applications in polymer composites: advancements, challenges, and future prospects. Journal of Materials Chemistry A. https://doi.org/10.1039/D5TA00982K
Yakoubi, S. (2025). Sustainable revolution: AI-driven enhancements for composite polymer processing and optimization in intelligent food packaging. Food and Bioprocess Technology, 18(1), 82-107. https://doi.org/10.1007/s11947-024-03449-2
Yu, K., Yao, Y., Zhang, W., Lu, L., Gao, Q., Zhang, P., & Sing, S. L. (2025). Defects, monitoring, and AI-enabled control in soft material additive manufacturing: a review. Virtual and Physical Prototyping, 20(1), e2588456. https://doi.org/10.1080/17452759.2025.2588456
Alqahtani, H., & Ray, A. (2024). Surface texture analysis in polycrystalline alloys via an artificial neural network. Measurement, 227, 114328. https://doi.org/10.1016/j.measurement.2024.114328
Liu, X., & Aldrich, C. (2022). Deep learning approaches to image texture analysis in material processing. Metals, 12(2), 355. https://doi.org/10.3390/met12020355
Ercetin, A., Der, O., Akkoyun, F., Gowdru Chandrashekarappa, M. P., Şener, R., Çalışan, M., ... & Bharath, K. N. (2024). Review of image processing methods for surface and tool condition assessments in machining. Journal of Manufacturing and Materials Processing, 8(6), 244. https://doi.org/10.3390/jmmp8060244
