Reindustrializing for the Age of AI
AI, supply chain constraints and evolving manufacturing technologies are reshaping reindustrialization efforts, highlighting the need for a forward-looking approach to rebuilding industry.
People often say that the biggest problem with computers is that they do exactly what we tell them to do, even when the result isn’t what we really want. Agentic AI now inverts that dilemma by allowing users to describe the desired result and leave the agent to calculate the best way to achieve it. Moreover, these agents can draw not only on resources within, say, a factory, but also on information from additional sources to evaluate and adapt their strategies over time as external factors change. This represents continuous improvement – a principle already deeply rooted in industrial quality-management strategies – that now operates autonomously and in real time.
AI processing can take place in the cloud or at the edge, on small, resource-constrained platforms. As factories incorporate more sensors, edge devices provide an ideal way to measure sensor readings and determine immediate responses. Processing, packaging and sending data to the cloud then provides the basis for longer-term analysis, helping identify trends and build the information sources AI agents need to optimize manufacturing and business processes.
The rapid adoption of AI in industrial edge devices such as smart sensors, PLCs, computing modules, and gateways puts the sector in competition with AI data center operators for access to resources such as high-end components, PCBs, and substrate materials. Explosive growth in the server market is diverting these resources from other sectors, and we should remember that industrial automation products account for about 20% of the global PCB market. While the sector currently represents only about 12% of the US PCB market, industrial applications account for more than 30% of European PCB production. Industrial PCBs account for about 6% of China’s total PCB production, as the country produces enormous volumes of boards, primarily for communication and computing equipment.
As AI server demand continues to consume high-end resources, limited access to advanced technologies could significantly hinder the West’s urgent drive to reindustrialize. Factories must adopt cutting-edge technology to deliver competitive quality and productivity. The offshoring boom has transformed manufacturing over the past two to three decades, and companies must establish new factories with full awareness of these changes.
While manufacturers initially pursued offshoring mainly to reduce costs, today’s push for reshoring aims to strengthen supply chain resilience and geopolitical independence. New approaches include split-sharing, which divides manufacturing operations across multiple locations worldwide to avoid geographical concentration. At the same time, China’s manufacturing landscape is changing as producers increasingly serve healthy and rapidly growing domestic markets profitably.
The current situation has evolved over more than a quarter of a century and will take more than a couple of years to rebalance. Although the West has no shortage of determination or financing, money always moves faster than people. Companies must hire and train suitable workers, and workforce development could become a bottleneck. We are approaching a time when humanoid robots could help address this challenge, and at least one company in Europe is currently planning a new factory that will incorporate AI-based principles and processes from the ground up. This strikes me as a bold and exciting move that highlights a direction Western industry could pursue as it strives to meet its objectives in the coming years. On the other hand, the hardware shortages we currently face will limit how quickly we can adopt advanced AI technologies, including humanoid robots, in new factories.
“Time is what keeps everything from happening all at once,” wrote author Ray Cummings in the short story The Girl in the Golden Atom. There is an undeniable truth at the heart of the joke. While we can implement small changes relatively easily and quickly, the major changes we now need are much more complex and will demand proper planning, a long-term perspective, and perseverance.
We must also consider the current circumstances: supply chain bottlenecks persist, new capacity takes time to come online, and establishing a new factory realistically takes two to three years. Meanwhile, technological change continues to accelerate.
Analyzing the origins of the current situation is unlikely to help guide the future. I remember participating in a working group after the dotcom crash of the early 2000s, tasked with identifying the causes and developing a way to anticipate similar events in the future. The crash had left the company with excessive inventory, and management wanted to avoid a repeat. We concluded that we could not have reliably predicted the situation, much less prevented it. I believe today’s circumstances are similar. Companies pursued offshoring for understandable and defensible reasons: technology made it possible, competitive pressures made it necessary, and the shift democratized high-tech manufacturing while sharing its benefits with emerging economies.
The next era of advanced manufacturing could begin here, and incorporating AI into every activity, from the factory floor to the front office, could unlock greater efficiency, delivery performance, profitability, and adaptability. It will be fascinating to watch this transformation take shape and a privilege for those of us here today to witness it and play a role in it.End of article content
Alun Morgan is technology ambassador at Ventec International Group (venteclaminates.com); alun.morgan@ventec-europe.com. His column runs monthly.