K-Line: Accuracy Is Decisive in Ship Performance Software

Box shipping: accurate data.  File Image / Pixabay.

Japanese shipping line Kawasaki Kisen Kaisha (K-Line) is to deploy machine learning technology across its 300-strong fleet, the company has said.

Applying machine learning methodology to improve vessels' performance is the watchword of US technology firm Bearing AI and since 2019, K-Line has been working with the firm on this issue.

While K-Line is a data collector, it is the accuracy of the data generated by Bearing's software that makes the difference.

"For many years, we have focused on the 'collection of high-quality data' and 'advanced data analysis technology' and have made various efforts to improve the accuracy of performance evaluation technology," the company said in a statement.

"As part of this, from the end of 2019, we have conducted a demonstration experiment on verification and evaluation of data analysis technology of AI provided by Bearing AI.

"As a result, we have confirmed that it is possible to evaluate operational performance with extremely high accuracy compared with existing performance evaluation technology."

It is that accuracy that points the way forward to further efficiency gains.

"We will not only accurately grasp the operational performance of each vessel operating in the actual sea area and use it for operational management, but also accurately evaluate various fuel reduction efforts," according to the company.

Bearing AI is a start up firm based in California's Silicon Valley co-founded by Dylan Keil and investor Andrew Ng.

Keill spoke to Ship & Bunker last week about their plans to introduce the company's services to the wider, shipping community.