How Manufacturers are Turning a Page on Automation and AI Due to the Pandemic

How Manufacturers are Turning a Page on Automation and AI Due to the Pandemic
How Manufacturers are Turning a Page on Automation and AI Due to the Pandemic
At the start of 2020, automation, Artificial Intelligence (AI) and robotics were all buzzwords evoking both technological marvel and fear that for each piece of innovation, the manufacturing floor would lose a human job.
Skip ahead a few months to the height of the pandemic: the world was facing a state of emergency, the manufacturing industry was in crisis, and the global supply chain was on the brink of destruction. With factories in danger of being shut down worldwide and social distancing mandates in place, the demand for new technologies—and AI, in particular—skyrocketed. Previously hesitant to adopt these technologies, manufacturers were forced to completely rethink their operations and processes.
In fact, when asked about priorities for their customers in 2020, industrial manufacturers said that implementing Industry 4.0 initiatives that improve quality and augment production yield and efficiency were key. And this is true for manufacturers of all sizes, not just the behemoths. The goal: more efficient, productive, competitive, and resilient factory floors that can survive and function in the "new normal."
The need for innovative processes when it comes to quality inspections is clear. This function is traditionally performed by human workers that, pre-pandemic, were often unavailable due to labor shortages. Now amidst the pandemic, the need is even greater. The pandemic has made workers either intermittently available, or unable to work at full capacity due to social distancing requirements.

How big of a problem is this? Statistics show quality inspections are being performed by some 35 million workers—that is roughly the population of Canada devoted just to performing this basic function on the manufacturing floor.  
Visual quality inspections powered by AI can be the answer to this dilemma, removing the barriers that typically slow technology adoption by being cost effective, easy to integrate, and not needing specially trained staff.
The breadth of applicability for visual quality inspections is vast—from inspecting raw materials coming into a plant and items coming down the production line, to finished goods being packaged or kitted, and boxes being palletized for shipment. Take the example of packaging, or kitting, of processed food into those ubiquitous freezer trays. It is common practice for inspectors to visually assess whether or not the right food is placed in the right spot in a tray, the portions are correct, and all spots are filled with food for a complete and nutritional meal. By augmenting this process with visual AI, manufacturers can inspect more product both efficiently and effectively, even with labor shortages and social distancing in place.
Another example of products that can benefit from AI-powered visual quality inspection are Printed Circuit Boards Assemblies (PCBAs). Because PCBAs are complex items to manufacture, they need multiple points of inspection—such as resistors, capacitors, solder points, HDMI ports, RAM or the USB controller—in order to be deemed “acceptable." This often requires AOI machines staffed with workers or workers doing visual inspection, both human-intensive.
But AI-powered visual inspection is just the tip of the powerful intersection of a much broader set of technologies, where Artificial Intelligence meets Internet of Things–or AIoT. Today’s production floors and industrial machines are equipped with dozens of inexpensive sensors and cameras gathering product data, as well as basic diagnostics from industrial equipment. Manufacturers are leveraging human-level AI that can extract actionable insights from that data at the compute edge, directly inside industrial machines.

In this case, AI works in tandem with human operators by learning what a "normal production run" is like for a given machine. This allows operators to manage multiple machines and spot issues in the production, which prior to AI, needed heavy human oversight and might have resulted in a production shutdown when defects were detected.  
Whether in cameras or embedded in sensors inside industrial machines, some are saying there is a quiet revolution taking place on factory floors. To prepare themselves for the next economic downturn or crisis, manufacturers are embracing Industry 4.0 technologies at pace never seen before.
The pandemic has had the unintended consequence of turbocharging a revolution that will make factories more competitive and resilient—way beyond 2021.  

About The Author

Dr. Massimilliano Versace is the CEO and Co-Founder of Neurala, where he continues to lead the world of intelligent devices after his pioneering breakthroughs in brain-inspired computing. He holds several patents and two PhDs: Cognitive and Neural Systems, Boston University; Experimental Psychology, University of Trieste, Italy.

Did you enjoy this great article?

Check out our free e-newsletters to read more great articles..