First published in Manufacturing North America
Smart technology continues to disrupt every sector of manufacturing, with terms such as Artificial Intelligence and Internet of Things (IoT) now being a part of everyday language rather than the buzzwords they were a few years ago. Even so, we are only just starting to understand the possibilities of smarter, smaller, cheaper and more connected technology.
The Fourth Industrial Revolution is appropriately titled given the global impact of connected technology and IoT.
And the pace of change is accelerating, which can be understandably overwhelming and daunting to manufacturers when deciding where to focus and what to prioritise.The good news is that there is a wide range and scale of potential benefits open to those who embrace the latest technologies. Great leaps in connectivity, along with huge advances in chip and sensor technology mean that testing and trialling new ways of working today is much more viable than it was in the past.
The hope of significantly improved productivity and profitability is, of course, hugely attractive to us all. And whilst innovation has felt like a buzzword recently, the most successful companies have been innovating for decades, if not centuries.
Jack Welch, former CEO of General Electric famously said,
“If the rate of change on the outside exceeds the rate of change on the inside, then the end is near”.
It’s likely you’ve heard these words before, yet they seem even wiser advice today than they did almost 20 years ago.
Finding the answers
Given the large scale of the challenge, it makes sense for technology providers, consultants and project teams to major on specific areas of smart manufacturing. One of my areas of focus over recent years has been manufacturing intelligence, even though I didn’t realise it would be that way.
Let me explain
We’ve always had an innovation programme at work, which is a fairly informal arrangement ensuring that we keep up to speed with the latest technologies and try to spot applications for them, so we can investigate their potential. This keeps everyone informed and motivated. And it works best when there is a real-world challenge to be solved, rather than just playing with the tech for the sake of it.
Aware that tracking the location of assets, tools and fixtures had been a long-standing headache across manufacturing, we set about to see if we could use emerging IoT hardware and connectivity to solve the problem.
Asset management is fairly well established, but systems relying on technologies such as RFID and barcodes are not without their limitations. The infrastructure required can be incredibly expensive, and we saw that location data was highly unreliable and inaccurate. Further, more recent options like GPS are currently only useful outdoors.
So, after reviewing all the technology available to us, we settled on Bluetooth to try and crack the asset location problem. We figured that unlike RFID, it is a modern technology that is actively being improved and developed, especially given it is vital for connecting consumer mobile devices and peripheral hardware.
A major barrier for using new technology and systems in manufacturing is the expense of installing new infrastructure. The cost for this is often larger than the actual technology which uses that infrastructure. So, we needed to be zero infrastructure, even down to using our own meshed network, rather than relying on internal IT and Wi-Fi.We created a fairly straightforward prototype consisting of two pieces of hardware, a detector and a tracker, plus a software interface to manage the system and track items. We were then fortunate to partner with Rolls-Royce Aerospace and a couple of other companies in the UK to trial the kit in actual facilities.
The initial results were pleasing. The asset tracking capability worked great in a real manufacturing environment, which was full of metal and interference that we thought might stop the kit from working. Motion and wastage time were reduced as assets were being found much quicker. Additionally, loss and scrappage levels also decreased.
Then it started to get even more interesting.
From tracking to intelligence
We started to realise that the data and analytics we were already collecting using the range of sensors in the hardware could go beyond location and could have multiple additional applications and benefits.
Here are just a few:
Asset utilisation and smoothing: how are duplicate fixtures and tools being used? Do they need to be used more evenly? This can help manufacturers to be more efficient with servicing and cleaning, and in determining a total number of fixtures and tools required for optimal operations.
Environmental measures: embedded sensors are able to collect temperature, humidity and air quality readings to show a full history of the conditions that an asset has been under. This is incredibly useful for composite material tracking.
Automated logic: this can be set up to alert individuals or teams when certain assets or groups of assets enter a particular area, haven’t moved for a length of time, become available, or any other rules-based trigger.
Damage and foreign object alerts: this function can detect impact from drops and collisions and raise notifications.
Predictive resource planning: integrating with ERP and other systems can ensure assets or groups of assets, such as tool kits, are in the right place ahead of time rather than being located at time of need.
Predictive maintenance: monitoring unusual vibration patterns and telemetry can help with scheduling maintenance before failure.
Workplace safety: the collected data can also raise an alarm when volatile chemicals come into dangerously close proximity with each other.
A couple of years further on, we continue to discover new applications for the system on a weekly basis. These are often derived from suggestions or questions that come through our website or at conferences and are from disparate sectors and companies. A selection of these includes: making commercial waste more efficient, tracking workers and equipment deep underground in mines, analysing insurance risk, and improving safety for lone factory workers.
Accessible for all
We’ve found it’s important to implement open systems that are straightforward to integrate with others. Whilst we chose to focus on Bluetooth and GPS, there is still value in working with other technologies (in our case RFID specialists) to provide a collaborative service that is overall better.The speed of development is also vital. Long and costly implementation times just won’t cut it in the IoT circles, so agility is required to allow partners and clients to contribute to steering the development roadmap.
Whilst the promise of Artificial Intelligence in manufacturing is to deliver self-optimising businesses, this still feels unreachable or distant for most. But by implementing some smaller scale manufacturing intelligence trials of IoT technology, many will start to see the benefits of the organisational and technological changes that will help them remain relevant and be successful for many years to come.
First published in Manufacturing North America