I am an avid shopper. In fact, I doomscroll Amazon more than Instagram and TikTok combined at this point. At the same time, I’m incredibly curious and am a firm believer that every product has a history. Even the mass-manufactured ones have a story to tell. In this day and age, that story, in many cases, is how AI-powered industrial vision systems scanned them, assessed their quality and approved their release to market.
AI is no longer a ‘trend’. It’s everywhere, used in every product we buy. It has become an intrinsic part of product engineering, development, quality assessment and even sales and marketing. And such pervasive use of AI warrants a conversation. That is the goal of this blog.
Let’s get started, shall we?
How AI is Redefining Quality in Manufacturing
Gone are the days of people only buying utility items at the cheapest possible rates. The modern buyer spends more and expects more.
We all love, support and buy Pinterest-worthy products. Aesthetics play a key role in our buying decisions. Unfortunately, that does increase manufacturing and Quality Assurance (QA) costs.
At the same time, we are less tolerant and largely unforgiving of defects. This is why manufacturers in the past have relied on vision systems. These systems rely on machine vision (the ability to analyse images) and coding that helps them define, identify and flag defects. The traditional systems require weeks of coding and testing. If, for example, you need these systems to identify new types of defects (which a human operator will be the one to identify first), you will need to spend another few weeks updating the programme.
Sounds like a costly and hectic affair, right? It definitely is.
Even with this effort-intensive process, unaccounted microscopic defects, hairline fractures in products and cracks are incredibly difficult to identify. These defects may seem unimportant initially, but anything as simple as rough handling during delivery, a very minor fall or human error in handling can increase costs significantly when such defects add up. Let’s also not forget that some industries (such as healthcare) cannot afford to market defective products (howsoever insignificant those defects may seem to an outsider).
AI not only solves every single one of these problems, but it also makes manufacturing scalable and operations smoother.
Here’s how it does that:
-
Deep learning algorithms can be trained within hours to identify new defects (it takes traditional vision systems weeks of coding to be able to deliver similar accuracy).
-
AI can flag new defects and even support root cause analyses by visualising and interpreting data (traditional vision systems can only identify the defects they are programmed to identify). In fact, AI algorithms can also be trained to predict defects and likely causes (this supports proactive maintenance of parts and machinery and reduces unplanned downtime).
-
AI can be trained to identify random and microscopic defects which even human operators may miss.
-
Because the entire QA process is AI-assisted, operations are more scalable and efficient. This saves manufacturers additional funds that they could spend on market research, product engineering and development and marketing.
How This Impacts Our Daily Lives
We’ve discussed a lot about how AI supports manufacturing. Let’s discuss why buyers like you and I can benefit from AI-powered industrial vision systems.
In the manufacturing world, there is a growing emphasis on human-robot collaboration. Research and analyses suggest that AI is now assisting real-time decision making and automation in many manufacturing units and factories.
This will make products more trustworthy, less expensive and more customisable. Over the next decade or so, AI will very likely become an integral part of the manufacturing process. This will enable buyers to demand defect-free products on time every time at a price they can afford.
However, this change is transformative and will require patience and collaboration as industries evolve and our buying demands change.
While chasing perfection is the key to attaining progress, we may need to reassess our buying priorities. Will we become intolerant towards hand-crafted products? Will artisans be able to cope (because they will very likely not be able to match) with increased standardisation and quality assurance of mass-produced products?
As a collective, we have a choice to make. At the same time, we must learn to adapt and accommodate both defect-free manufactured products and homemade and hand-crafted speciality products. At the very least, we need to start becoming more mindful of our purchases and how we approach the things we buy.