From Manuals to Minutes: How Rockwell Automation is Fusing Decades of Expertise with GenAI

Imagine being an engineering assistant responsible for keeping a high-tech factory running. Suddenly, a machine stalls. In the past, someone like Mangleswaran Mahalingam would have to hunt through hundreds of pages of physical manuals, searching for an obscure error code. If the manual failed, he’d have to track down a veteran colleague who might have seen the problem years ago. It was a slow, trial-and-error process that cost precious time. Today, that entire cycle has been condensed into a few seconds.

Since October 2025, the team at Rockwell Automation’s Singapore facility has undergone a radical shift. They are now using the GenAI-Powered Maintenance Copilot, a bespoke AI assistant that acts as a bridge between decades of human expertise and modern machinery. This isn't just a generic chatbot; it is trained specifically on the collective knowledge of Rockwell’s workforce, manufacturing software data, and technical manuals.

The Brain Behind the Shop Floor

Built on the Microsoft technology stack, specifically using Azure OpenAI in Microsoft Foundry, this tool utilizes OpenAI’s GPT-5.4-mini model. For technicians like Mahalingam, it means their primary tool is no longer a thick binder, but a tablet or smart AR goggles. Instead of asking around for a senior technician, he can type a natural language query and receive an instant, accurate solution.

Rockwell, the Wisconsin-based industrial giant, is the backbone for many global operations—from airport luggage routing to soft drink bottling. Their Singapore factory, however, has become the ultimate testing ground for how AI can preserve the "tribal knowledge" that often disappears when veteran engineers retire. With some experts boasting 30 years of experience, capturing that tacit knowledge in a digital brain is the only way to ensure the next generation can lead effectively.

Real-World Gains: Downtime and Dollars

When a machine triggers an error now, the AI doesn’t just spit out a code. It organizes data by symptom, cause, and reaction. It tells the technician the most likely culprit—be it a faulty sensor or a worn gear—based on historical data and expert logs. It even highlights if the same issue has cropped up on other production lines, ensuring a standardized approach across the entire facility.

According to Singapore plant director Li Wang, the results are undeniable. The facility has seen a 33% reduction in machine downtime. Furthermore, by being more surgical with repairs, Rockwell’s internal tracking shows that spending on servicing and spare parts has dropped by 25%. It turns out that when you know exactly what is wrong, you stop wasting money on the wrong fixes.

Slashing the Learning Curve

The human impact is perhaps even more impressive. Mahalingam recalls the anxiety of his early days in 2021, when the sheer complexity of the factory felt overwhelming. It took him over a year to feel comfortable working night shifts alone. Now, thanks to the Copilot, that onboarding window has been shattered.

Bob Buttermore, Rockwell’s chief supply chain officer, notes that the time it takes for a new hire to become competent in troubleshooting hundreds of machines has dropped from nine months to just three. That 67% improvement in "time-to-competency" is a game-changer for a sector facing a global talent shortage.

A Global Lighthouse for Innovation

The Singapore plant is more than just a factory; it was recently named a World Economic Forum global lighthouse site. The accolades are backed by hard data: a 43% increase in units per person-hour and a 35% reduction in defects. These gains aren't just from the Maintenance Copilot, but from a suite of AI tools, including real-time quality assurance systems and predictive maintenance models.

Patrick Dey, VP of data and AI innovation at Rockwell, emphasizes that this was made possible by leveraging the full Microsoft ecosystem—including Azure AI Search for reasoning and Azure’s enterprise-grade security for compliance. This ensures that while the AI is smart, it is also secure and scalable.

Scaling the Future

Rockwell isn't keeping these secrets to themselves. The Singapore facility serves as a blueprint for the company's 25+ global plants and their international clientele. Plans are already in motion to roll out the Maintenance Copilot to manufacturing hubs in Twinsburg (Ohio), Poland, and Mexico.

As Chris Nardecchia, Rockwell’s CIO, puts it, the goal is a future where AI is "in the loop" for every task. By testing these innovations in their own "living lab" in Singapore, Rockwell is proving to its customers that the factory of the future isn't a distant dream—it’s already online, learning, and solving problems faster than ever before.