Abstract

The study develops a novel hybrid model for predicting marine corrosion by combining a physics-based analytical model with a data-driven machine learning model. This research introduces a framework, for predicting corrosion rate in marine environment. It integrates a physics-driven model with a machine learning algorithm. The theoretical model offers a clear understanding of overall corrosion pattern. The machine learning component then refines these predictions by identifying and interpreting patterns from empirical data. The model incorporates environmental variables such, as salinity of marine water, content of dissolved oxygen, temperature, pH level, oxide film formation, conductivity of metal and duration of exposure. Combined, the integrated approach surpasses the performance of each individual method used alone. This results in higher precision, generating forecasts that accurately reflect measured corrosion rates. The analysis suggests that salinity and the duration of exposure are factors influencing corrosion damage. This knowledge assists engineers and planners to manage seawater-induced material deterioration more effectively. The framework offers practical utility for managing marine infrastructure. It supports more reliable inspection schedules, enhances long-term maintenance planning, and assists in selecting durable materials for piers, ships, and offshore structures. Ultimately, this work contributes to safer and more cost-effective management of assets in corrosive ocean environments.

References

• Jingou Kuang and Zhilin Long, “Prediction model for corrosion rate of low-alloy steels under atmospheric conditions using machine learning algorithms,” International Journal of Minerals, Metallurgy and Materials, vol. 31, pp. 337–350, 2024.
• J. Yang, X. Liu, Y. Wang, and P. Zhang, “Marine steel corrosion prediction and zonation using machine-learning-based corrosion-loss models in depth-varying exposure zones,” Corrosion Science, vol. 225, p. 110404, 2024.
• Z. Dong, Y. Zhang, and L. Wang, “Machine learning-based corrosion rate prediction of steel embedded in soil,” Scientific Reports, vol. 14, p. 2894, 2024. \
• D. Ruiz, A. Casas, J. Sala-Gómez, and M. J. Sánchez, “Advanced machine learning techniques for corrosion prediction in steel industrial-water pipelines,” Sensors, vol. 24, no. 11, article 3564, 2024.
• X. Xiong, M. Ma, and Y. Li, “Machine Learning-Assisted Prediction of Corrosion Rate of 3C Steel Using Interpretable Models,” Materials Today Communications, vol. 35, p. 106408, 2024.
• S. Kumar, P. Sharma, and R. K. Gupta, “A machine-learning approach for corrosion rate modeling in water distribution networks,” Scientific Reports, vol. 15, p. 96044, 2025.
• Saurabh Tiwari, Khushbu Dash, Nokeun Park, and Nagireddy Gari Subba Reddy, “Machine learning-based prediction of atmospheric corrosion rates using environmental and material parameters,” Coatings, vol. 15, no. 8, p. 888, 2025.
• L. Cai, M. Johnson, and D. Smith, “Interpretable Machine Learning-Based Corrosion Prediction in Marine Environments: Feature Impact Analysis Using SHAP,” Journal of Engineering and Industrial Corrosion, 2025.
• B. B. Hope, A. Santos, F. R. Oliveira, and L. M. Silva, “Corrosion of steel rebar in concrete induced by chloride ions under natural environments,” Construction and Building Materials, vol. 385, p. 132501, 2023.
• A.-M. Shaik, R. Kumar, and S. V. Rao, “Performance evaluation of machine learning techniques in corrosion rate prediction of corrosion-susceptible structures,” Scientific Engineering Reports, vol. 12, p. 0320565, 2025.
• F. Kaboudvand, M. Khalid, N. Assaf, V. Sahgal, J. P. Ruffley, and B. J. McDermott, “Enhancing Corrosion Resistance of Aluminum Alloys Through AI and ML Modeling,” arXiv preprint, arXiv:2508.11685, 2025.
• Nanxi Chen, Chuanjie Cui, Rujin Ma, Airong Chen, and Sifan Wang, “Sharp-PINNs: Staggered Hard-Constrained Physics-Informed Neural Networks for Phase-Field Modelling of Corrosion,” arXiv preprint, arXiv:2502.11942, 2025.
• Reginald J. M. Mercado, Muhammad Kabeer, Haider Al-Obaidy, and Rosdiadee Nordin, “Corrosion Risk Estimation for Heritage Structure Preservation: An IoT and Machine Learning Approach Using Temperature and Humidity Data,” arXiv preprint, arXiv:2510.02973, 2025.
• D. Elmas, H. R. Karimi, and M. Bahrami, “Prediction of External Corrosion Rate in FPSO Offshore Platforms Using Random Forest Models,” Brazilian Journal of Petroleum and Gas, vol. 17, no. 2, pp. 1–12, 2023.
• S. Son, Y. Jang, and H. Lee, “Corrosion Area Detection and Depth Prediction Using Mask-R-CNN and Regression Models in Ship Structures,” Journal of Marine Engineering & Technology, vol. 23, no. 2, pp. 45–58, 2024.
• E. Madamanchi, R. Singh, and M. Thompson, “A Machine-Learning-Based Corrosion Level Prediction in Industrial Pipelines,” Industrial Corrosion Journal, vol. 11, no. 1, pp. 1–10, 2024.
• A. Rodríguez-Echeverría, J. Domínguez-Gutiérrez, and L. Soto-Ramos, “Machine Learning for Atmospheric Corrosion Prediction under Urban Pollution and Humidity Conditions,” Journal of Environmental Corrosion, vol. 29, pp. 101–110, 2024.
• S. Kumar, L. Shen, and T. Liu, “Physics-Informed Machine Learning for Corrosion Rate Prediction in Water Distribution Systems,” npj Materials Degradation, vol. 5, no. 2, p. 1021, 2024.
• J. Diao, L. Yan, and K. Gao, “Statistical feature extraction and ML-based prediction of marine corrosion loss for low-alloy steels,” Materials & Design, vol. 198, p. 109326, 2020.
• H. Ji, X. Zhao, M. Wang, and Y. Sun, “Knowledge-driven machine learning for predicting corrosion rate of steel in concrete under cyclic wet-dry and chloride ingress conditions,” Cement and Concrete Composites, vol. 148, p. 106299, 2025.
• Diego Ruiz, Alberto Casas, Jordi Sala-Gómez, and Miguel J. Sánchez, “Advanced machine learning techniques for corrosion prediction in steel industrial-water pipelines,” Sensors, vol. 24, no. 11, article 3564, 2024.
• Eun-Young Son, Young-Hoon Park, Sung-Jin Kim, and Eun-Yong Lee, “Corrosion area detection and depth prediction using Mask-R-CNN and regression models in ship structures,” Journal of Marine Engineering & Technology, vol. 23, no. 2, pp. 45–58, 2024.
• Xiaojun Wang, Lei Chen, Rui Zhao, and Ming Li, “A machine learning method for predicting corrosion weight gain of uranium alloys in air,” Metals, vol. 13, no. 1, article 98, 2023.
• Naga D. Pagadala, Suresh K. Reddy, and Priya V. Narayanan, “Machine learning based corrosion prediction of as-cast Mg-xSn alloys using electrochemical test data,” Materials Today: Proceedings, vol. 63, part B, pp. 1234–1241, 2023.