In a world where manufacturers face constantly shifting volumes, quality demands, and delivery expectations, the need for flexibility has never been greater. The newest MPD Podcast episode brings together two industry leaders, Fastems, expert in factory automation, and Ingersoll Rand, a global compressor manufacturer, to explore how artificial intelligence can be applied today in real manufacturing environments.

Fastems CEO Mikko Nyman and Jani Mannila, Manager of Production and Method Engineering at Ingersoll Rand, discuss how their long-term collaboration culminated in a new, intelligent factory line built for flexibility and future-ready AI.

“For this investment, we decided to choose Fastems as our main partner because they can provide all the data handling. Our responsibility is then the actual machining process, where we think we are the best in the world. And if Fastems is the best in their business and we are the best in ours, then when we combine these two and build a new line together, we can truly achieve great things,” says Jani Mannila.

Flexibility Drives the Need for Smarter Manufacturing

As manufacturing shifts away from mass production to high-mix, variable-volume production, agility becomes a survival skill. Fastems defines flexible manufacturing as the ability to run any job on any machine, anytime, while maximizing spindle hours and minimizing lead times.

“Flexible manufacturing responds to the changes in the world, which are becoming faster and increasingly complex. It's increasingly more difficult to predict what happens in manufacturing, what should be produced and when. That’s why dedicated manufacturing lines are not viable investments anymore. You need increased flexibility in your production so that the system can, within certain boundaries, create any kind of parts in any kind of batch sizes needed,” says Mikko Nyman.

For Ingersoll Rand, that meant investing in a new-generation Flexible Manufacturing System (FMS) integrated with automation, tool management, coolant recycling, and data systems. The outcome? Reduced manual work, fewer hazards, and consistently high quality, even at low volumes.

The 4-Step Program to Enabling AI on the Shopfloor

Fastems and Ingersoll Rand outline a practical 4-step framework to bring AI into real-world machining.

Step 1: Automation & Repeatable Process

The foundation of smart manufacturing begins with reliable automation. Ingersoll Rand started by minimizing human interference in machining to ensure constant quality and higher spindle utilization.

“For us, the first step is of course to make sure that our processes are reliable and they work. We know most of the time what is going to happen, but that is only based on our own experience. It’s not measurable data. That is why we started to think, is there software or any kind of solution we could integrate into our manufacturing processes to get that data?” says Mannila.

Step 2: Integrate Surrounding Processes

With machining processes automated, Ingersoll Rand turned their focus to surrounding tasks that still relied heavily on manual work. They began integrating systems to automate tool measurement, part measurement, and other operator-dependent steps, making the production line more accurate and less reliant on human intervention.

“In the new system, we have integrated a lot of human-related tasks... Now that you can automate them, it has reduced the time where we need our operators. It’s also made the system more accurate, as the machines are doing most of the work, measuring the tools, measuring the parts...” says Mannila.

Step 3: Collect and Analyze Data

In complex manufacturing environments, data plays a central role in enabling smarter operations. At Ingersoll Rand, data collection starts from real-time part measurement during the machining process. This information is used to adjust the process immediately when needed and to track tool power and tool life. Statistical analysis helps detect recurring tolerance issues, allowing further refinement of machining programs and machine setup.

Fastems emphasizes that meaningful data must flow not just from machines but also across systems, ranging from ERP and CAD/CAM to quality and tool management. Their solutions, like MMS Dashboard and Factory Cockpit, are designed to unify this data and turn it into actionable insights for production planning, control, and long-term development.

Step 4: Act and Adjust Based on Data

This final step pushes toward true AI behavior, where insights derived from data enable real-time process adjustments. While this is still evolving, both companies emphasize that the building blocks are already in place.

About the Podcast

The MPD Q, A & I podcast series, part of the Manufacturing Performance Days, dives deep into the role of AI in creating new digital business models, advancing technologies, and leading sustainable transformation. Through candid conversations with leading professionals at the forefront of AI transformation, the series offers listeners first-hand insights into the future of AI-powered manufacturing.

Listen to the full podcast episode:
In Spotify
https://open.spotify.com/episode/6gGa84iBi6lmVD7IvmKyaN?si=ba31ee9f57124f8e

Apple Podcasts
https://podcasts.apple.com/us/podcast/mpd-q-a-i-flexible-manufacturing-in-practice-how/id1810859680?i=1000706916042