NTT Data & Hyster-Yale deploy AI for assembly checks

NTT DATA and Hyster-Yale Materials Handling have introduced an artificial intelligence system for assembly quality checks in manufacturing. The system is now in use at Hyster-Yale Materials Handling's plant in Berea, Kentucky.

The deployment applies what the companies describe as physical AI to a critical assembly workflow, using sensor data to track whether production steps have been completed and parts fitted before a product moves to the next stage.

Designed and developed by NTT DATA at the facility, the setup combines vision sensors, on-site edge AI processing and analytics. NTT DATA worked with Hyster-Yale Materials Handling and Archetype AI to adapt a model that compares assembly activity with expected production steps and flags deviations.

The project focuses on quality assurance during assembly rather than after products leave the line. By checking installation and process completion as work progresses, the system is intended to identify problems on the factory floor before they move further into production.

Factory use

The companies described the approach as an early industrial assembly use of physical AI embedded directly into production workflows. They said running the system locally at the edge keeps data processing on-site rather than sending it elsewhere for analysis.

That local processing is central to their case for faster implementation. According to the companies, early results showed deployment timelines shrinking from months to weeks compared with older methods.

Manufacturers have been seeking AI systems that can operate in complex industrial settings, particularly in inspection, process control and production reliability. In that context, the Berea deployment offers a practical example of AI applied to a specific assembly task rather than used as a broad planning or reporting tool.

For Hyster-Yale Materials Handling, which makes Hyster and Yale lift trucks, the work is part of a broader push to apply digital tools across global manufacturing operations. The aim is to maintain quality standards while supporting reliable production at scale.

Barbara Binda, Director of Global Manufacturing Innovation at Hyster-Yale Materials Handling, outlined the company's view of the project.

"Our confidence in physical AI continues to grow, and we're starting to see the countless benefits that AI can bring to our global manufacturing operations," said Barbara Binda, Director of Global Manufacturing Innovation at Hyster-Yale Materials Handling. "Working with NTT DATA allows us to leverage how physical AI can help our production teams maintain high-quality standards and deliver the most reliable products to our clients."

Broader trend

The announcement reflects a broader shift in industrial technology, as manufacturers test AI models against live operational data from cameras, sensors and machinery. The appeal lies in using those systems to observe physical processes in real time and respond before defects or missing steps become larger production issues.

Edge computing has become an important part of that model because it keeps processing close to the data source. In manufacturing settings, that can reduce delays and make it easier to fit AI tools into existing operational technology environments.

NTT DATA said the work with Hyster-Yale Materials Handling builds on a longer relationship between the two companies. It also pointed to further efforts to make manufacturing processes more adaptive and repeatable through AI in production environments.

Shahid Ahmed, Global Head of Edge Services at NTT DATA, said the project shows how the technology is moving from concept to use inside working factories.

"This deployment shows what physical AI looks like in real production environments, not as a concept, but with tangible impact on the factory floor," said Shahid Ahmed, Global Head of Edge Services at NTT DATA. "By combining real production data with physical AI models at the edge, we're helping leading manufacturers like HYMH deliver high-quality products, support frontline workers and apply AI in ways that deliver real-world outcomes."