The next meeting of the ASM Materials Informatics Technical Committee will be held on Wednesday, May 13, at 1:00 PM EDT.
The meeting will feature the talk "Smart AI-Driven Materials Science Analytics for Corrosion Analysis," presented by Thomas Considine, DEVCOM Army Research Laboratory, and Nicholas Josselyn and Biao Yin, Worcester Polytechnic Institute.
Anyone with an interest in artificial intelligence or corrosion analysis is welcome to attend the meeting. To request the online Zoom link, please contact scott.henry@asminternational.org.
Abstract
Traditional corrosion testing has long relied on manual data collection, subjective visual ratings, and fragmented record-keeping - all in environments hostile to both materials and researchers. This presentation describes how the U.S. Army's DEVCOM Research Laboratory has developed an AI-driven platform to modernize and automate this process. The work encompasses deep learning models for corrosion classification and segmentation, as well as domain adaptation methods that use indoor test data to predict outdoor corrosion performance - potentially eliminating the need for some costly, years-long field exposures. These capabilities are integrated into MOSS (Materials Open Science Software), a platform that combines an iPadOS field data-collection app, a centralized data repository, and a web portal for AI-assisted visual analytics. Together, these tools represent a significant step toward faster, more objective, and more cost-effective materials validation - with broad implications for defense applications and the coatings industry.
The presentation is based on an article that the presenters and other coauthors contributed to the upcoming ASM Handbook, Volume 26, Artificial Intelligence for Materials Science and Engineering.
------------------------------
Scott Henry
Director of Content and Publishing
ASM International
------------------------------