Initial Data Analysis Supports New Corrosion Detection Tool by ABS

Image courtesy of ABS.

Powered by an artificial intelligence (AI) platform, results from an initial application of the Corrosion Detection digital solution by the American Bureau of Shipping (ABS) (Houston, Texas, USA) show it has identified corrosion with 98.9-percent accuracy.  The ABS AI Corrosion Detection tool continues to refine its fidelity as more images are analyzed to train the algorithm.

“Corrosion is one of the biggest maintenance challenges in the maritime industry, with the total cost estimated at between $50 billion and $80 billion annually,” says Kash Mahmood, senior vice president for digital solutions at ABS.  “This application of advanced machine learning technology to assist with tank coating and corrosion management is unique to ABS, and now we are able to demonstrate that it is highly effective in real world applications. ABS is leading the industry with deployment of digital technology with practical tools that help shipowners drive efficiencies in their operations.”

According to CBS, a client recently deployed the Corrosion Detection tool to examine two vessels to develop an accurate assessment of their coating conditions and the presence of corrosion. Analysis reveals the tool graded the imagery with 98.9-percent accuracy. The findings are based on the AI tool’s analysis of more than 600 images of a client’s ballast water tanks.

As a structural-focused digital solution available to owners, the ABS AI Corrosion Detection system utilizes an AI algorithm that analyzes images and video files to deliver prediction scores and percentages for tank coating breakdown and corrosion. The digital solution allows operators to benchmark and monitor coating breakdown and corrosion growth over time to assist with vessel maintenance between surveys. By leveraging the power of machine learning’s automated pattern identification, clients receive proactive alerts of structural defects such as coating failures, fractures, and deformations without the costly and intrusive aspects of traditional structural inspections.

According to ABS, the application will continuously improve its performance as more images are uploaded and evaluated, further refining the precision of its detections.

Source: ABS,