We have been told “Smart Manufacturing is coming” for many years and buzzwords like the Industrial Internet of Things (IIOT) and Digital Twins have been tossed around as the next new thing for some time now. Intuitively this makes sense – we rely on our connected digital devices to help us through our daily lives. And yet, factories have been slow to embrace these changes because the landscape is confusing and not well defined. So what is Smart Manufacturing, and why are people excited about it? Furthermore, what is still holding us back from making good on the promise it brings?
Smart Manufacturing means bringing the elements of smart technology – sensing inputs, computing power, always-on connectivity, artificial intelligence, and advanced data analytics – to the traditional production process. Used collectively, these technologies should help teams unlock new opportunities to accelerate development, reduce waste, and increase transparency of the supply chain.
Connected sensors and input devices – things like tracking devices in trucks, environmental monitors in factories, or cameras on the assembly line – make up the Industrial Internet of Things (IIoT) and are designed to monitor steps of the supply chain, machine health, and factory conditions. Often these devices consist of a sensing element, an efficient microprocessor brain, and a radio or other way to communicate the information back to a centralized hub. With the continued miniaturization of technology, these devices are now small, efficient, and cheap enough to be installed everywhere on the factory floor and throughout the supply chain.
But just gathering the data is not enough. To be useful, the information collected by the sensing devices needs to be able to be parsed and analyzed leading to actionable insights. Here is where artificial intelligence and big data come in. With enough data, machine learning and AI platforms can be trained to identify anomalies in production quality or standard procedures. On the assembly line, this can mean speeding up the issue discovery process and fixing problems before they escape into the field.
Building a comprehensive picture of the assembly process and each device being built can lead to new opportunities in manufacturing. Imagine a problem has been found on a shipping device. It is unclear if the issue was present throughout development or if it occurred recently. With historical data collected at every step of the way in the smart factory, it would be easy to isolate devices that use similar lot codes as the one that failed. Records from the date of manufacture can be analyzed to root cause the failure quickly. This can be done to some extent today, but with smart systems, all of this could be done in real time.
If enough data can be collected about a given unit, a digital twin can be created alongside each device. Information about the product’s design, specific component origins, particulars of the factory conditions, assembly process, and test results can be linked to a physical product so that a complete history can be accessed from anywhere in the world in a matter of seconds. Without access to a device, engineers from across the globe can look at the same data set to troubleshoot and solve issues.
While it may seem like manufacturers should be jumping at the opportunity to incorporate these smart technologies into their factories, there are a number of challenges with getting this right. Turning what has traditionally been a manual process like an assembly line into a smart factory does not happen all at once and the manufacturers themselves may not be incentivized properly to make the investment.
Operationally, manufacturers will need to invest in both the hard infrastructure but also the talent required to operate, maintain, and update the smart ecosystem. This is not an easy calculation when margins are already low and they don’t always have the best tools available to evaluate all the options. Complicating matters are concerns about security and privacy that come with all networked systems. No brand wants their top secret projects leaked to the public ahead of a launch.
In addition, the smart devices themselves will need to be reliable, durable, and accurate, but more importantly they will also need to play nicely with a wide variety of standard and home-grown data platforms. As yet, there are no one-stop shops where these sensors and controllers can be acquired and no platform that everyone uses.
For smart systems to be truly effective, all vendors in the supply chain will need to be connected together. This means that in addition to qualifying vendors on traditional metrics like speed, communication, and quality, you will now need to budget time to validate the tech stack and link APIs to make sure the data platforms can communicate in both directions.
At Instrumental, we are making a manufacturing optimization platform that helps you realize the promise of smart manufacturing. We collect product data from your assembly line and correlate your test results so you can find anomalies in real time. Our machine learning algorithms can also find issues you didn’t even know to look for so you can find and fix more issues faster. And we work with you and your partners to make sure all the data is consolidated in a single place.