A year-end look at how AI, simulation, edge computing, and a rapidly evolving workforce are reshaping digital transformation strategies and real-world operations on the plant floor.
As 2025 winds down, conversations about digital transformation have shifted. The term no longer appears on every trade show banner or vendor brochure, but according to Engineering.com senior editor Mike Ouellette, that’s not because manufacturers have moved on. It’s because digital transformation is no longer optional, or novel.
“Digital transformation isn’t new at all,” Ouellette said on a recent episode of the Ctrl+Alt+Mfg podcast. “Humans have this need to put ideas in buckets. Digital transformation is just the latest bucket we’ve created to describe the digitized functional advances manufacturers have been implementing for 30 years.”
From the rise of programmable logic controllers (PLCs) in the 1980s to Industry 4.0 and now the emergence of artificial intelligence (AI)-driven Industry 5.0, manufacturers have long been modernizing incrementally. What’s different today, Ouellette argues, isn’t the intent, but the available tools.
The democratization of compute technology
Across the industry, manufacturers now have access to advanced compute power, cloud infrastructure, low-cost sensors, robotics and AI capabilities that simply didn’t exist even a decade ago.
“The computational power we have available now, at scale, is the real game-changer,” Ouellette said. “Before, you’d need a server farm on-site to do what you can now do in the cloud through AWS or Oracle. The democratization of compute is what makes this era different.”
Large manufacturers have long deployed simulation, robotics and advanced automation, but today even small and mid-sized manufacturers can deploy digital tools affordably and modularly. Technologies that once belonged only to high-budget aerospace or automotive plants are now within reach for job shops and contract manufacturers.
The individual nature of digital transformation
Despite the accessibility of tools, digital transformation remains a highly individualized process. Two factories making the same product may have entirely different digital roadmaps, driven by variations in legacy equipment, workforce skills, workflow constraints and business goals.
“Everyone has access to the same technologies. What’s different is the problems they’re trying to solve,” Ouellette said. “When you get to the shop floor, the issues are custom. You have to mesh the technology with your people, your expertise and your pain points.”
This fragmentation explains why digital transformation is rarely a single sweeping project. Instead, it comprises a series of targeted deployments — predictive maintenance here, connected supply chain visibility there, a digital twin for one line or cell before expanding to others.
And because the landscape evolves so quickly, many leaders still struggle simply to keep up.
How digital twins, robotics are creating real value
While hype can be distracting, the value-producing technologies today are familiar:
- Predictive maintenance and machine data analytics
- Digital twins and simulation
- Cloud enterprise resource planning (ERP), manufacturing execution systems (MES) and product lifecycle management (PLM) built on common data layers
- Modern robotics and cobots
- Edge computing for legacy equipment.
Digital twins stand out for Ouellette as a particularly high-impact tool.
“Digital twins are incredibly powerful. You can run simulations for quoting, new product specs or determining whether you need to buy new machinery,” he said. “You’re no longer guessing; you’re testing.”
Robotics has also become dramatically more accessible. Ouellette recalled earlier trade shows where only massive industrial robots dominated the floor.
“Now you can walk into a booth and see a cobot the size of a human hand placing a screw,” he said. “That capability used to cost hundreds of thousands of dollars. Today you can rent it.”
The manufacturing workforce challenge
Technology is transforming roles from engineering through operations. The greatest shift, Ouellette says, is in mindset.
“It’s the execution jobs that are changing most, not the strategic ones,” he explained. “You have to be a technical native now. You need to understand why a system works the way it does and then adapt to new tools quickly.”
AI is accelerating that shift. Engineers who once spent hours running computer-aided design (CAD) simulations now evaluate thousands of AI-generated design variations. Test and validation engineers no longer gather machine data manually. Application programming interfaces (APIs) do that work.
Yet resistance to AI remains a cultural hurdle. Ouellette sees similarities to early automation adoption.
“AI is there to augment the human, not replace the human,” he said. “It’s good at things humans are terrible at — fast calculations, massive datasets, reasoning on huge information sets. It makes people better at what they do.”
Why pilots stall and how to break through
One of the biggest challenges manufacturers face is escaping “pilot purgatory.” Many organizations test technologies, see limited early impact and never scale.
Ouellette identifies several root causes:
- Legacy equipment that wasn’t designed for connectivity
- Data challenges and lack of internal analytics expertise
- Budget constraints after years of supply chain and economic disruption
- Uncertainty about return on investment (ROI), or even how to calculate it.
“Manufacturers are expected to invest in digital initiatives after years of cost increases and disruptions. They’re scraping the bottom of the barrel for capital, so they start small,” he said. “But that’s not a bad thing. You learn what works, build ROI and fund the next project.”
Cybersecurity: The underrated risk
As factories connect more equipment, cyber risk grows. But the biggest vulnerabilities are often in the oldest assets.
“We’re still seeing operational technology get hacked because someone put an edge device on a 20-year-old machine with no cybersecurity built in,” Ouellette said.
Even basic password hygiene remains a problem.
What’s next: AI agents, edge and simulation
Looking three to five years out, Ouellette sees three technologies driving the most change:
- AI agents trained on manufacturers’ own data
- Digital twins and simulation at every scale
- Edge computing that brings legacy equipment into the digital ecosystem.
“Manufacturing is one of the best-positioned industries to take advantage of AI, especially agentic AI,” he said. “The possibilities are endless.”
Start small, win big with digital transformation
Digital transformation is delivering on its promise—for the companies that start early, stay focused, and build momentum incrementally.
“If you’re not doing it at scale, start small, then do another small thing, and another. Next thing you know, you’ve done a big thing,” Ouellette said.
In 2025, the buzzword may be fading, but the transformation is only accelerating.
The Ctrl+Alt+Mfg podcast
Make sure to check out the first few episodes of the Ctrl+Alt+Mfg podcast, where hosts Gary Cohen and Stephanie Neil discuss a range of digital transformation insights:
Ep. 1: Resetting and Rethinking Manufacturing
Ep. 2: Uniting Disparate Data With John Lee, Matrix Technologies
Ep. 3: Rethinking OT Security With Leah and Jeremy Dodson, Piqued Solutions
Ep. 4: Making Digital Transformation Real With Alicia Lomas, Lomas Manufacturing
Ep. 5: Reducing MES Project Risk With Ryan Crownover, Vertech
