TT DATA and Hyster-Yale Materials Handling (HYMH) deployed a physical AI system that integrates vision sensors, edge AI, and analytics directly into a critical assembly workflow at HYMH’s manufacturing facility in Berea, Kentucky. The co-developed system analyzes production activity in real time, validates assembly steps, and flags deviations before products advance to the next stage of manufacturing.
NTT DATA developed the system with HYMH and technology partner Archetype AI. The companies adapted a physical AI model to process sensor data locally at the edge and compare assembly activity against expected production sequences. The model verifies that required parts are installed and production stages are completed, allowing quality issues to be identified during assembly rather than after products leave the factory floor.
The companies said early results reduced deployment timelines from months to weeks compared with legacy techniques. Local processing allows production data to remain on-site while supporting faster analysis and system rollout. NTT DATA and HYMH plan to explore additional applications of physical AI across HYMH’s global manufacturing operations as manufacturers increasingly combine AI models, edge computing, sensors, and operational technology systems.
• Deployment location: HYMH manufacturing facility in Berea, Kentucky.
• Architecture: Vision sensors, edge AI processing, physical AI models, and advanced analytics.
• Primary use case: Real-time validation of critical assembly processes and detection of production deviations.
• Processing model: AI inference and production data analysis run locally at the manufacturing site.
• Deployment timeline: Early results reduced implementation time from months to weeks compared with legacy techniques.
• Technology partner: Archetype AI contributed the physical AI model adapted for the manufacturing environment.
• Expansion strategy: NTT DATA and HYMH plan to evaluate how the architecture can support repeatable quality assurance across additional manufacturing operations.
“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, NTT DATA, Inc. “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.”
🌐 Analysis: The deployment illustrates how physical AI architectures can extend enterprise AI infrastructure beyond centralized cloud and data center environments by placing inference, sensor processing, and decision support close to industrial equipment. The model also reflects growing convergence between edge computing and operational technology, as NTT DATA and other infrastructure providers develop platforms for deploying AI models across geographically distributed factories, warehouses, and industrial sites.
🌐 We’re tracking the latest developments in AI infrastructure, edge computing, and physical AI. Follow our ongoing coverage at: https://convergedigest.com/category/ai/
| Profile: NTT DATA | |
| Headquarters | Tokyo, Japan |
| Leadership | Yutaka Sasaki, President and CEO, NTT DATA Group |
| Scale | $30+ billion business and technology services organization |
| Global Reach | Operations and experts across more than 70 countries; serves 75% of the Fortune Global 100 |
| Core Technology | AI Edge Computing Cloud Digital Infrastructure |
| Key Services | Enterprise AI, cloud services, cybersecurity, connectivity, data centers, applications, consulting, and industry solutions |
| Physical AI Architecture | Vision sensors + edge inference + physical AI models + advanced analytics |
| Manufacturing Milestone | Deployed physical AI for real-time assembly validation at Hyster-Yale’s Berea, Kentucky manufacturing facility |
| Strategic Objective | Integrate AI across IT and operational technology environments to support distributed, data-driven industrial operations. |