By Stephan Pottel, Manufacturing Strategy Director, EMEA, Zebra Technologies
“What’s in a name?”, Shakespeare’s Juliet famously asked. In the world of manufacturing, the answer is: a lot. The terms “automation” and “AI” are on the lips of many leaders, yet their meanings have become so confused that they now risk undermining the very progress they are meant to represent.
For some, automation can mean streamlining processes, but for others, it is about embedding deep operational intelligence or orchestrating complex workflows. Similarly, AI can refer to a predictive algorithm or an adaptive learning system. But while a rose by any other name might smell as sweet, mislabelling a solution can risk costly mistakes, downtime, and even reputation, while buzzwords can mask the technology’s true potential to make work better every day.
From Simple Steps to a Bigger Dance
For many business leaders, the first step into automation is all about tackling individual tasks. It’s a move that Zebra’s research with Oxford Economics shows is widely seen as essential for businesses to stay competitive and handle change.
This is the ground floor of automation. It’s the classic idea of using a machine or a piece of software to do a single, repetitive job. Think of a robotic arm on a production line or software that processes invoices automatically. The real value here is getting things done with speed, consistency, and accuracy, especially for tasks that don’t change much. It’s the foundation for a more efficient business.
Taking things a step further is what’s known as workflow automation, an approach used by about four in ten manufacturers. Here, the view widens from a single action to a whole sequence of them. Several automated jobs are linked together to make a whole process run smoothly. This gets rid of the clumsy manual steps and data re-entry that slow things down and cause errors. The result is a much smoother operation and a real boost in productivity.
Despite being different in scale, both of these approaches have something important in common: they are based on following strict rules. They are experts at handling the expected, designed to carry out their instructions perfectly every time. They automate the ‘doing’, but they cannot handle anything uncertain, manage unexpected changes, or learn from experience.
From Guessing the Future to Understanding the Present
So, what about AI? It is not one single thing, but a whole range of abilities that allows systems to learn, think, and adapt. For most in the manufacturing world today, AI is seen as a tool for advanced analysis. This is the practical side of machine learning, where it’s used to sift through huge amounts of data to make very good predictions. As research by Zebra and Oxford Economics, shows businesses are using it to:
- manage their stock better
- predict when a machine needs maintenance, and
- improve the quality of their products.
This type of AI automates the hard work of data analysis to help people make smarter decisions. While it’s brilliant at spotting that one thing is related to another, it doesn’t truly understand why.
Bringing Intelligence into Focus with Machine Vision
Then there is a more advanced and transformative type of AI which is all about adaptive learning. Using a sophisticated method known as Deep Learning, it moves beyond neat and tidy data. It learns to make sense of the complex, real world of a factory floor, including interpreting 2D and 3D images from cameras and sensors.
At a basic level, a camera on a production line automates inspection, following pre-set rules to spot known defects. But when adaptive AI is added, it can now understand what is happening. This allows a system to learn from its own experience, finding subtle patterns and problems that it was never programmed to look for. It’s a ‘virtuous loop’ of constant improvement, where the system gets smarter with every item it inspects.
The focus shifts from asking, “What do we predict will happen?” to “What can we learn from what is happening right now to make the whole process better?”. This is the leap from simply automating analysis to creating a form of digital thinking, enabled by machine vision.
From Words to a Real Advantage
The words “automation” and “AI” are neither written in water nor in stone. What the manufacturing industry means by them has evolved. Business leaders should strive for a deeper knowledge of their different levels and applications, from tackling individual tasks to workflow automation, and from data analysis to deep learning. This more sophisticated understanding will empower decision-makers to choose the right tool for the right job and so create a real advantage for their business.
