The Rise of Physical AI: How Hitachi and Daikin are Pioneering Intelligent Factories
Tokyo – The convergence of artificial intelligence and physical systems is rapidly transforming manufacturing, and a collaboration between Hitachi and Daikin Industries is at the forefront of this revolution. Hitachi is focusing on developing “physical AI” – systems where AI directly controls robots and industrial equipment – and is leveraging data from across its diverse businesses to create advanced and, crucially, safe systems.
AI-Powered Diagnostics: A Game Changer for Manufacturing
Daikin and Hitachi began a trial operation in April 2025 to implement an AI agent designed to support equipment failure diagnostics in factories. This initiative, centered at Daikin’s Rinkai commercial air conditioner factory in Sakai, Osaka, aims to combine Daikin’s operational technology expertise with Hitachi’s advanced IT capabilities. The goal? To improve productivity and minimize downtime.
The AI agent works by converting factory equipment drawings into a “knowledge graph” readable by generative AI. This AI is then trained on maintenance records and other operational data, enabling it to diagnose faults with accuracy comparable to, or exceeding, that of experienced Daikin maintenance engineers. A proof-of-concept experiment demonstrated the AI’s ability to identify the cause and countermeasures for equipment faults within 10 seconds, with over 90% accuracy.
Beyond Diagnostics: The HMAX Ecosystem and Future Applications
This collaboration is part of Hitachi’s broader HMAX strategy, launched in January 2026, designed to accelerate social innovation across industries. Under HMAX for Factories, the Daikin partnership represents a key step in providing AI-driven solutions for complex industrial environments.
The potential applications extend far beyond air conditioner manufacturing. Imagine AI agents proactively identifying potential failures in automotive production lines, optimizing energy consumption in chemical plants, or ensuring the smooth operation of critical infrastructure. The ability to predict and prevent equipment failures translates directly into cost savings, increased efficiency, and improved safety.
Building Safe and Reliable Physical AI Systems
Hitachi’s approach emphasizes safety, and reliability. By combining multiple open models with its accumulated data, the company aims to build AI systems that are not only powerful but also robust and trustworthy. This is particularly important in industrial settings where even a minor malfunction can have significant consequences.
The initial collaboration between Daikin and Hitachi began in 2017 with the establishment of a next-generation production model utilizing Hitachi’s Lumada IoT platform. This earlier work focused on digitalization, comparison, and analysis of worker skills to support predictive maintenance and manufacturing process optimization, laying the groundwork for the current AI-driven diagnostics project.
The Evolution of AI in Manufacturing: From Prediction to Prescription
The shift from predictive maintenance to prescriptive maintenance – where AI not only predicts failures but also recommends specific actions to prevent them – is a key trend shaping the future of manufacturing. The Daikin-Hitachi collaboration exemplifies this evolution, offering a glimpse into a future where AI agents are integral to the operation of intelligent factories.
Did you know? Generative AI is being used to create knowledge graphs from existing factory documentation, making it easier for AI agents to understand and diagnose complex equipment issues.
FAQ
Q: What is “physical AI”?
A: Physical AI refers to AI systems that directly control physical devices, such as robots and industrial equipment.
Q: What is the benefit of using AI for equipment diagnostics?
A: AI can diagnose faults faster and more accurately than traditional methods, reducing downtime and improving productivity.
Q: What is Hitachi’s HMAX strategy?
A: HMAX is Hitachi’s initiative to accelerate social innovation across industries through advanced technologies, including AI.
Q: When did Daikin and Hitachi commence their collaboration?
A: Daikin and Hitachi began collaborating in 2017, with the current AI-driven diagnostics project starting in April 2025.
Pro Tip: Investing in data infrastructure and creating comprehensive knowledge graphs are crucial steps for successfully implementing AI-powered diagnostics in manufacturing facilities.
Aim for to learn more about the latest advancements in AI and manufacturing? Explore our other articles on industrial automation and the future of work. Subscribe to our newsletter for regular updates and insights.
