By: Virginia Shram
February 10, 2022
This is the first article in our weekly series, Eyes on Industrial AI, where we explore the multifaceted world of industrial artificial intelligence and automation. Let’s uncover the innovative applications of these cutting-edge technologies specifically as they relate to manufacturing and supply chain management in today’s era of Industry 4.0, including any challenges they encounter.
Over the past couple centuries of civilization, humans have wildly overestimated where we’d be in terms of technology in the modern age.
Years of anticipatory sci-fi buildup, from Trekkie transporters to the Jetsons’ insta-meal robot, have hinted to an achievable future teeming with flying vehicles and personal robot assistants. This technological optimism goes beyond entertaining fiction, too, since we are routinely promised utopias of self-driving cars and the elusive, perfect AI-generated list of Netflix recommendations based on your previous watches (sigh… one day it’ll be accurate).
So, if artificial intelligence truly lived up to the hype, where are all our jetpacks?
Interestingly, for all our cultural obsession with the soaring limits of AI, we hardly understand what artificial intelligence truly does or how it’s incorporated into societal and manufacturing infrastructure. In other words—
Aside from wildly overestimating humanity’s progress, we have also wildly underestimated our technological achievements when it comes to applying them.
The big reason why is that this AI stuff gets real complicated, real quick. And without an awe-defying hook, like identifying Alzheimer’s cells 100 times faster than human researchers, it’s tough to work through the technical details in order to understand the impact. After all, what is automation and artificial intelligence but building something that has far greater capabilities than the human brain designing it?
All that we tend to hear about are the big-ticket, inspirational projects at the front of the race, like neural implants or self-driving cars. These projects, aside from being very cool, are amazing historical markers of technology at the edge of innovation. Most of us, however, are a few steps back from the edge, but still just as much pushing the whole movement forward through participation.
Just because most working examples of artificial intelligence in manufacturing don’t look like Boston Robotics canines doesn’t mean they aren’t just as critical to the development of the whole field of automated intelligence.
This series will explain the actual implications of AI as an emerging technology, from big-picture applications like aerospace exploration and lifesaving medicine, to equally critical applications like parts and packaging production in factory assembly and supply chain management.
Artificial intelligence is a term so huge that it encompasses many different scientific methodologies, from statistics to biology to philosophy. In manufacturing, it tends to encompass the fields of big data, machine learning, and high-level automation tasks (known as full automation).
Here’s a short list of some terms that are widely used (sometimes interchangeably) to cover the field of industrial automation:
Over the course of this series, we’ll go through each term individually, but for now, try to think of all of these topics as overlapping subgenres of artificial intelligence.
Surely it’s a completely different technology that identifies dead cancer cells in a split second than that which controls a driverless delivery van throughout unfamiliar routes, right? Isn’t one just “sorting,” and the other “thinking”?
… not really.
Here’s the thing — smart technology is not an either-or scenario.
Whether it’s machine learning algorithms or natural language processing, all artificial intelligence lies upon a spectrum of smart technology, including levels of independent automation.
In most cases, introducing AI applications into your technical framework won’t immediately mean hyper-intelligent face recognition robots. In most cases, good AI applications may be instantaneous mathematical models that can assess real-time scenarios far faster than human workers ever could.
In other words, it’s not really an issue whether to adopt artificial intelligence or not — unless you’re a philosopher or sci-fi writer concerned with theoretical and emotional consequences. The question of industrial AI is to what extent are your operations scalable technologically? How is your business modernizing at a similar rapid pace to that of Industry 4.0? In the same way that you seek to minimize waste from production, seek change through technological opportunism.
Automation isn’t lesser to AI, but they both exist along a spectrum that defines the manufacturing cutting edge of adaptive intelligence.
A realistic look at the capabilities of Industrial AI requires a critical view of the infrastructure within which the AI system will operate.
No matter whether you’re making top-of-the-line robotics systems or are “just” a single-product manufacturer, industrial AI is a critical tool that you can wield to spur actionable growth and efficiency.
For an accurate look at the latest in artificial intelligence, McKinsey Analytics’ study, “The State of AI in 2021,” references global market changes in the field of automation. As expected, it reports that companies are overall more invested in AI applications than they were the previous year, but the gains from AI adoption were “unlikely to compensate for the pandemic era’s global supply-chain challenges” (4). This quote shows that despite the general increase of AI applications, unexpected supply chain issues still derailed production.
This is important because, while it’s true that AI can solve problems we haven’t even yet conceived of, it’s not a deus-ex machina type magical solution that will inevitably solve every problem. An optimistic, critical understanding of industrial AI acknowledges the limits and challenges of technological growth spurts. This series will highlight the particular manufacturing details that come with these challenges, and so we will pay attention to some of the trending frameworks employing AI in industry.
According to the same report, “Most companies—whether they are high performers or not—tend to use a mix of cloud and on-premises platforms for AI similar to what they use for overall IT workloads. But the high performers use cloud infrastructure much more than their peers do: 64 percent of their AI workloads run on public or hybrid cloud, compared with 44 percent at other companies” (7).
Even if you are dubious that your company will benefit from automation — especially if you’re stubbornly anti-AI — you have to admit that the global leaders are moving to cloud computing services to satisfy higher processing demands required of AI applications. The most successful of these global companies make the most thorough, extensive use of processing power and artificial intelligence. This supports the theory that Industry 4.0 isn’t here to stay, it’s here and it’s evolving faster by the minute.
Will you keep up?
In Eyes on Industrial AI we’ll cover more industrial AI news and predictions in the future, including delving deeper into some of the intellectual and practical consequences of full industrial automation and how it provides us a glimpse at the future of the Smart Factory in Industry 4.0.
Here are some questions and topics we’ll cover in the near future, so stay tuned for more, including:
… and much more!