You are here

Corrosion Modeling Analytics and Machine Learning to Promote Corrosion-Informed Design to Reduce Ship Maintenance


OUSD (R&E) CRITICAL TECHNOLOGY AREA(S): Advanced Computing and Software; Advanced Materials; Sustainment


OBJECTIVE: Develop (1) protocols algorithms to transform various raw data formats into information-rich features for machine learning (ML), and (2) software and modeling tools for ML that will automatically detect patterns in data; and learn and augment from experience and corrosion–informed models the ability to predict optimal materials selection and/or corrosion control measures to reduce Navy ship maintenance.


DESCRIPTION: A number of Navy ship classes face growing maintenance delays and maintenance costs. Some solutions resort to cannibalization of parts: moving them from one ship to keep another one operational. This is a critical issue as Navy ships are getting fewer steaming hours because of growing maintenance delays and costs. Maintenance delays have resulted in some ships deferring maintenance. Over time this situation has resulted in worsening ship conditions and increased costs to repair and sustain ships. In some cases, maintenance has been deferred to the point where ships have been decommissioned several years ahead of their planned service life. With increasing computer capabilities, growing materials databases, increasing computational capabilities, the growing use and power of ML and artificial intelligence, digital engineering can reduce acquisition timelines and cost, permit more rapid system upgrades, and streamline maintenance. In addition to verified corrosion models in relevant operational environments, failure analysis, inspection reports, documented ‘lessons learned’, and the results of past maintenance practices can be incorporated into the materials database.


The challenge of digital engineering for DoD is attaining knowledge-based integration of data sufficient to decide lifecycle issues. Key elements of digital engineering are developing and compiling materials databases and developing relevant corrosion models that can predict materials behavior and operational life in platform systems operating in marine environments. ML is a powerful subset of artificial intelligence (AI) for systems to learn from data, pattern identification, and decision making. Application of ML tools can enable characterization of materials and informed-corrosion behavior in new ship design and inform Navy Maintenance personnel about options for cost-effective materials or corrosion control methods to lessen future ship maintenance. A key challenge in applying ML algorithms to materials science data is that data comes in many formats. Determining how to featurize and utilize different materials data formats so that prior data can be used as training data for ML algorithms can be difficult. Feature engineering, including extraction, transformation, and informed selection, is critical for improved ML accuracy and increase Fleet operational availability.


PHASE I: Define and develop a concept/approach/framework for feature engineering tools to extract critical information related to corrosion and other degradation pathways (e.g., physical, strength, fatigue resistance, etc.) from multiple formats. Key features may also include material properties, chemistry, and processing variables. Include in the concept/approach/framework appropriate identification classifiers and interactions. Assemble verified corrosion models and other descriptive terms for different materials. Develop a Phase II plan. In a Phase I option, if exercised, demonstrate the feasibility of the proposed concept/approach to provide labeled data output for corrosion/corrosion control options.


PHASE II: Develop, demonstrate, and validate a materials database for supervised (e.g., support vector, neural networks) and unsupervised learning algorithms (e.g., cluster analysis) use for corrosion/corrosion control and life prediction. Ensure that the collective database is able to identify prioritization of features whether it be structural, chemical, and physical properties or AM-related processing-microstructure-property phenomena relative to corrosion phenomena.


PHASE III DUAL USE APPLICATIONS: Transition optimized computational/informatics handling engineering tools for commercialization in ML utilization through original equipment manufacturers (OEMs) or other partnering agreements. Demonstrate the technology to DoD warfare centers/production facilities. The design tool is focused on application in a marine environment so offshore structures such as oil and gas platforms could benefit.


Dual use applications could include aircraft, land vehicles, materials processing entities.

Commercialization of this technology may be realized via success in predicting materials service life in marine and modified marine environment in ship systems.



  1. Witten, Ian; Frank, Eibe; Hall, Mark A. and Pal, Christopher J. “Data Mining: Practical Machine Learning Tools and Techniques (Fourth Edition).” Morgan Kaufmann Series in Data Management Systems, Elsevier, 2000.
  2. Ling, Julia, et al. “Machine Learning for Alloy Composition and Process Optimization”. (Proceedings of ASME Turbo Expo 2018 Turbomachinery Technical Conference and Exposition.)
  3. LaQue’s Handbook on Marine Corrosion, 2nd ed,. Electrochemical Society Monograph Series, D.A. Shifler, Ed., John Wiley & Sons (June, 2022)
  4. D.A. Shifler, “Designing for Affordable Corrosion Control in Marine Environments", Proceedings of 2023 DoD Corrosion Prevention Technology and Innovation Symposium, August 14-17, 2023, Tucson, AZ.
  5. S. Singh, S. Mohan, “AI is Driving Digital Transformation in Engineering”, Control Engineering,


KEYWORDS: Machine learning; corrosion; modeling; digital engineering; design; maintenance; materials dataset

US Flag An Official Website of the United States Government