Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

Artificial intelligence: The revolution fueling the revolution

Is AI fueling the next industrial revolution?

A futuristic digital lightbulb hovers over cupped hands

[Adobe Stock / Kolapatha]

Profile photo of Joel Scutchfield

General Manager of SMT and Semicon Business Operations at Koh Young Technology · Koh Young America, Inc.

Sometimes it feels like we’re talking the digital talk, but not walking the digital walk, right? It was close to 10 years ago that we started talking about Industry 4.0, digital transformation, and smart manufacturing. In fact, it was not just talk, it was the must-have signage on every booth at every trade show, the must-see presentation at every conference, and the ‘next big thing’ that would revolutionize productivity and quality in our industry.

But it didn’t, and truth be told, the electronics manufacturing industry has always been more evolution than revolution.

SO WHY NO DIGITAL TRANSFORMATION—OR AT LEAST NO DIGITAL REVOLUTION?

Unlike digital transformation, the answer to that question is not difficult. Digital transformation is simply too complex, too overwhelming, too hard, and has barely even been defined. What’s more, I really don’t think the ingredient technologies were there a decade ago to make it happen. At the end of the day, digital transformation requires connectivity to get the data out of all the machines, synthesis and standardization of that data to derive any insight, and the ability to process that data and insight to feed back to the line and create real value in terms of improved performance, be that quality or efficiency.

Yes, some solid progress has been made. The CFX connectivity standard is allowing the entire SMT line to be connected and, against the odds, the entire industry seems largely happy to adopt this single standard. But we’re still on that long arduous journey, and that still leaves some legacy equipment isolated, as well as many processes that exist after the SMT line, like box build, cabling, packaging, etc.

As software evolves to connect processes and create feedback loops, the first and most successful example has probably been the connection and intelligent communication between SPI (solder paste inspection) and solder paste printing. These simple closed loops serve as an example of the potential of digital transformation, proving ROI (return on investment) is a straightforward use case with measurable costs and measurable value.

Another example is the synthesis of data from multiple inspection machines on multiple lines and even in multiple facilities to drive factory- and company-wide process improvement and improve staff efficiency in the face of talent shortages. Having said this, it still feels like we’ve got a long way to go, and perhaps that’s because we have a lot of data terabytes—and I mean a lot—flowing from the machines on the line every hour of every day! What’s more, we’re trying to figure out what to do with it all, what data to prioritize, and what data will give us the most improvement in performance—and hence, the most bang for our buck.

Right now, it feels less like a revolution where the old ideas are beheaded on the guillotine and more like the kind of evolution that took us from horse-drawn carriages to the motor vehicle.

CAN AI SPARK A REVOLUTION? 

We are in the decade of AI. AI stocks are garnering insane valuation, with companies like Open AI leading the pack. AI startups seem to be a part of every VC fund’s portfolio, and it seems that almost every day we hear of a new application for AI that is not just in development, but in beta or even full deployment. What’s more, those tech titans are all betting the proverbial farm on AI.

Rest assured, this is a revolution and it’s the revolution fueling the fourth industrial revolution. AI is accelerating the smart factory that we have been waiting for. It is the key ingredient technology that was missing for years, and is what can take all that connected data and derive insight and value at line speeds.

AI might not be the whole picture. AI wouldn’t function without all the work that has been done to connect machines, all the work that has been done to create standards to harmonize the data from those machines, and of course, all the work that has been done to collect quality, accurate data that feeds the machine learning needed to create intelligent systems and processes. I suspect it is the part of the jigsaw puzzle that was missing and the part that will deliver a digital dividend so we can stop talking about the smart factory and start to deliver that smart factory.