Shipbuilders turn to AI and robots to build 'smart yards'

Home > Business > Industry

print dictionary print

Shipbuilders turn to AI and robots to build 'smart yards'

HD Hyundai Mipo Dockyard in Ulsan holds a demonstration showcasing collaborative robots and advanced automatic welding techniques on a 50,000-ton petrochemical carrier under construction. [HD HYUNDAI]

HD Hyundai Mipo Dockyard in Ulsan holds a demonstration showcasing collaborative robots and advanced automatic welding techniques on a 50,000-ton petrochemical carrier under construction. [HD HYUNDAI]

 
At HD Hyundai Mipo Dockyard in Ulsan, sparks scattered across the deck of a 50,000-ton petrochemical carrier as robotic arms glided along rails, welding seams that once demanded a human’s steady hand. 
 
The scene offered a glimpse into a potential future for Korean shipbuilders: yards where artificial intelligence (AI) and machines, not muscle memory, drive the rhythm of construction.
 

Related Article

 
Deck welding is one of the most difficult tasks to automate, as a deck's surface is uneven, especially with obstacles such as pipes. But during a mid-October demonstration showcasing collaborative robots and advanced automatic welding techniques, the robots performed the job skillfully in front of the press.
 
Robotics has become a top priority for shipyards. Unlike most manufacturing industries, shipbuilding involves extensive outdoor operations that make automation difficult. So far, robots have been used mainly for indoor work, such as cutting steel plates or welding flat, straight panels. But shipyards like HD Hyundai Mipo are expanding their use of robots and refining their precision. In fact, the company expects to raise the automation rate for deck block welding from 58.6 percent to about 80 percent. 
 
HD Hyundai Samho Heavy Industries runs an Automation Innovation Center that operates and trains more than 80 welding robots. 
 
This year, it became the first in the world to apply collaborative robots to the production of large curved hull blocks, or the sections at the front and rear of ships, made from bent steel plates. Traditionally, skilled “plate benders” have performed this task manually, heating and shaping metal using experience and instinct.
 
Turning that intuition into data is the next challenge.
 
“Training AI with plate benders’ knowledge requires quantifying their experience,” said Ryu Cheol-ho, a professor of shipbuilding and mechanical engineering at Inha Technical College. “But shipyards still lack enough of that data.” 
 
The Diden 30, a quadruped robot designed to operate autonomously in high-risk environments inaccessible to humans, features a foot-shaped leg structure and magnetic feet. [KAIST]

The Diden 30, a quadruped robot designed to operate autonomously in high-risk environments inaccessible to humans, features a foot-shaped leg structure and magnetic feet. [KAIST]

 
Samsung Heavy Industries, meanwhile, signed an agreement on Oct. 24 with Rainbow Robotics to develop shipbuilding robots, starting with AI-equipped welding machines and expanding to mobile dual-arm and quadruped robots. The company is also working with startup Diden Robotics to adapt spiderlike welding robots whose magnetic feet let them cling to steel hulls for real shipyard use.
 
Additionally, Korea’s three major shipbuilders — HD Korea Shipbuilding & Offshore Engineering, Hanwha Ocean and Samsung Heavy Industries — are betting heavily on building smart shipyards. They aim to use robots and AI to improve productivity and eventually turn high-tech shipyards into exportable platforms. Tech companies, including Nvidia and Naver, have joined the effort to develop the AI platforms that will serve as the “brains” of these shipyards. 
 
Hanwha Ocean is investing 160 billion won ($111 million) between 2024 and 2026 to convert its Geoje yard into a “smart yard” and plans to apply the technology to its U.S.-based Philly Shipyard.
 
Samsung Heavy Industries unveiled the industry’s first automated design platform for shipbuilding and offshore engineering, the “S-EDP,” on Oct. 29. The company aims to connect data across all processes — design, procurement and production — by 2030 to achieve full digital transformation.
 
HD Hyundai introduced its “visible shipyard” digital twin system in 2023 and is in the process of expanding it into a fully connected yard that uses big data and AI to predict and optimize production by next year. 
 
By 2030, it plans to operate an “intelligent autonomous shipyard” that increases productivity by 30 percent and cuts construction times by 30 percent.
 
Apprentice workers at the training academy inside Hanwha Philly Shipyard in Philadelphia receive hands-on instruction in welding and other shipbuilding techniques. [KANG TAE-HWA]

Apprentice workers at the training academy inside Hanwha Philly Shipyard in Philadelphia receive hands-on instruction in welding and other shipbuilding techniques. [KANG TAE-HWA]

 
On Oct. 31, Naver and Nvidia announced plans to jointly develop a physical AI platform for real-world industrial sites. The platform will combine Nvidia’s Omniverse and Isaac Sim 3-D modeling and robotics tools with Naver Cloud’s digital twin and robotics technologies. It will simulate shipyard environments virtually and train AI models in those conditions. 
 
A Naver Cloud spokesperson said the platform would allow customized software development based on shipyard-specific data.
 
“The physical AI platform will allow customized software development based on each shipyard’s data,” the Naver Cloud spokesperson said, adding that the company plans to collaborate with HD Hyundai by using its data.
 
Lee Eun-chang, a research fellow at the Korea Institute for Industrial Economics & Trade, said smart shipyards could help solve a key challenge in overseas expansion — the shortage of skilled workers — by reducing reliance on human expertise. 
 
“It may even become possible to export the shipyards themselves,” Lee added.


This article was originally written in Korean and translated by a bilingual reporter with the help of generative AI tools. It was then edited by a native English-speaking editor. All AI-assisted translations are reviewed and refined by our newsroom.
BY LEE SU-JEONG [[email protected]]
Log in to Twitter or Facebook account to connect
with the Korea JoongAng Daily
help-image Social comment?
s
lock icon

To write comments, please log in to one of the accounts.

Standards Board Policy (0/250자)