Hyundai creates units for software-defined factory deployment

Hyundai Motor Group has created internal units to accelerate a software-defined factory (SDF) strategy and to procure robot components, according to The Korea Herald and AzerNEWS. The group appointed Alpesh Patel to a new leadership role overseeing SDF work and established a new office for robotics component sourcing headed by Soh Hyun-seong, the outlets report. Earlier public announcements cited by The Korea Herald say Hyundai aims to produce 30,000 humanoid robots annually by 2028 and to deploy 25,000 robots across Hyundai Motor Co. and Kia manufacturing plants; the report adds that Boston Dynamics' Atlas will begin parts sequencing work in Georgia in 2028 and expand into parts assembly by 2030. Editorial analysis: This is an operational step toward large-scale humanoid deployment in auto manufacturing.
What happened
According to The Korea Herald, Hyundai Motor Group has created units dedicated to advancing its software-defined factory (SDF) strategy and to procuring robotics components. The Korea Herald, citing industry sources, reports the group created a leadership position for the SDF initiative and appointed Alpesh Patel, who serves as chief innovation officer at Hyundai Motor Group Innovation Center Singapore. The outlet also reports that Hyundai set up a new office for robotics component procurement headed by Soh Hyun-seong, formerly head of strategic planning at Beijing Hyundai. The Korea Herald further cites earlier public announcements that Hyundai aims to produce 30,000 humanoid robots annually by 2028 and to deploy 25,000 robots across Hyundai Motor Co. and Kia plants. The report says Boston Dynamics' Atlas will handle parts sequencing at Hyundai Motor Group Metaplant America in Georgia beginning in 2028, with parts assembly work expected to follow around 2030.
Editorial analysis - technical context
Companies pursuing a software-defined factory approach pair centralised software control, machine vision, logistics orchestration, and advanced robotics. Industry-pattern observations note that integrating humanoid robots like Atlas into existing factories differs from fixed-arm automation because humanoids operate in human-centric layouts and require robust perception, balance control, and task generalization. Observers following the sector will watch whether mass-production of humanoid hardware and modular components reduces unit cost and shortens integration schedules compared with bespoke automation projects.
Context and significance
Automakers have been investing in software-first factory designs and increased automation to improve flexibility and customization. If manufacturers scale humanoid production, suppliers and parts divisions will face larger, more consistent demand for sensors, actuators, power systems, and specialised subassemblies. Reporting by The Korea Herald notes Boston Dynamics sought collaboration with Hyundai Mobis to mass-produce six key Atlas components, which illustrates how production partnerships may emerge between robot OEMs and traditional automotive suppliers.
What to watch
- •Whether Hyundai and Boston Dynamics announce formal production agreements or component contracts, and the named suppliers involved.
- •Pilot performance metrics from the Georgia Metaplant America facility on task throughput, error rates, human-robot safety incidents, and retrofit costs.
- •Boston Dynamics' manufacturing ramp cadence and unit-cost trajectory for Atlas hardware.
- •How Hyundai phases deployments across regions named in reporting, such as India and South Korea, and any regulatory or labor disclosures tied to automation rollouts.
Scoring Rationale
This is a notable operational development for industrial robotics and factory automation, with concrete production and deployment targets that matter to manufacturers and suppliers. It is not a frontier AI research release but has practical implications for deployment, procurement, and integration.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
