The manufacturing sector faces huge challenges, as do the supply chains connected to it. A series of manmade and other crises have contributed to a perfect storm of risk: soaring energy, fuel, and raw materials costs; looming recession; war in Europe; transport disruption; skills shortages; Brexit (for UK businesses); environmental and sustainability needs; and the long tail of the COVID-19 pandemic, which has left many firms struggling.
So, factories can’t afford unplanned downtime and expensive machine failures, as these may snowball through the sector, damaging clients and their own dependent customers. Ultimately, failures could have an impact on all of us in shortages and empty shelves, forcing up prices even higher.
One company believes that artificial intelligence and machine learning (AI and ML) are the answers, but not in the way that other businesses in the hype cycle do – we are surrounded by vendors making similar claims. This one uses an unusual approach: sensors that detect vibrations, sound, and electrical faults on the front end, then it deploys AI and ML behind the scenes.
With complex machinery running 24/7 on production lines, sometimes in safety cages or lights-out factories, faults often don’t come to light until it’s too late: when a critical component fails. But unless you can see inside every machine and monitor each widget – which is expensive and impractical – how can you tell if a lurking problem may lead to catastrophic failure?
The answer is by detecting minute changes in the way machines sound or vibrate, says Israeli entrepreneur Saar Yoskovitz, co-founder and CEO of New York-headquartered machine-health unicorn, Augury (“Machines talk, we listen”).
Founded in 2011, with full launch and first funding in 2014, the company now employs over 400 staff worldwide and works across a broad range of manufacturing sectors, such as food, drink, paper, forest products, chemicals, and pharmaceuticals. Blue-chip clients include Pepsico, Nestlé, Hersheys, Colgate-Palmolive, Heineken, Danone, Canfor, and Bausch Health, among others.
The first spark of inspiration came from an unlikely source: national service in Israel. Yoskovitz says:
My co-founder [CPTO Gal Shaul] was in the Navy and I was in the artillery. For both of us, our lives depended on very big machines. And in those scenarios, you become very ‘intimate’ with the machines, in that with every small squeak or crackle you can really understand what's going on. So, it was intuitive for us to say, ‘If I can hear something is wrong, why can't we make machines and computers understand that as well?’
At university, my studies focused on speech recognition using AI and machine learning algorithms. What we do today is very similar. We listen to the machine, we take an audio wave, and we try to find meaning inside it. But instead of searching for words in speech recognition, we search for patterns of malfunctions.
We pick up vibration signals, but we also have magnetic sensors that can detect electrical faults in the motors and the drives, temperature sensors, and so on. But the driving force was, if we can hear that something is wrong, then we can train a computer model to do that too.
Unique insights from unique data
So, Augury is all about data, but of an unusual and overlooked sort. By gathering vast amounts of data about what healthy machines and components sound like, any minor deviation can be flagged as a potential problem. Yoskovitz adds:
That’s exactly right. If you have a pump in your factory, we don't need to build a new model for your specific one, because we've seen over 20,000 pumps before. We know exactly what cavitation or cracked bars sound like. We have over 200 million machine hours that we've monitored, and all that data is sitting in our cloud.
We can use it to further refine our algorithms to make them more accurate and create new levels of insights, and we have pre-built models for all these different types of machines. From the very first moment when the sensor is installed on your machine, we can tell you what's wrong with it. As a result, within three months, all our customers pay back our full annual programme, and they start expanding.
Bold claims, and the company’s website claims 7 x ROI or more in some industries.
But a number of recent reports have revealed that, in general, smart factory initiatives often fail, because of the difficulties of scaling up from a successful pilot, upgrading old facilities, and finding the right skills, plus change management shortfalls. For example, pre-COVID Capgemini research suggested that just 14% of large-scale smart manufacturing projects succeed. Why do firms struggle with making their factories more efficient?
Acknowledging the problem, Yoskovitz says:
There's this term ‘pilot purgatory’, where pilots go to die. Our market today isn't good at proving value in a small pilot then scaling it up, that’s right. But we were able to crack the code of proving value very, very quickly.
But then how do you scale out, globally, with large companies? It turns out that the key to unlocking it is time to value: the faster you can provide value to the customer, the faster they can go to their manager and get more investment and scale.
We have the fastest time to value possible, and the reason for that is that we have a full- stack solution: we have our own hardware, we manage the connectivity, we run proprietary AI diagnostics on the data set, and we provide professional services around change management, reliability, and more.
The value of machine health
Which brings us to the company’s name. An augury is a sign or omen of the future (something that does, or does not, augur well). To what extent is Augury in the business of that definitive Industry 4.0 value-add: predictive, rather than reactive, maintenance? Yoskovitz says:
That's exactly what we do, it’s where we started. And over time, as we work with large customers like Pepsico, we understand that machine health is not a maintenance problem. It's much greater than that: it's a supply-chain risk-management problem, it's a workforce empowerment problem.
With [snack manufacturer] we helped them manufacture a million pounds of their product that they wouldn't have been able to do, because we detected and fixed a problem that increased their uptime, and therefore increased their capacity.
And we have the largest cement manufacturer in North America. They’ve predicted we're going to increase their capacity by two full factories. So, machine health is not just ‘Hey, you need to replace this bearing’, it's ‘Hey, you can avoid spending $800 million on building two new factories’, not to mention the environmental impact of that.
So, we’ve relabelled this category from predictive maintenance to machine health. But that’s just one side of the coin. The other side is process health, because we know that mechanical health impacts on the quality of the product the machine manufactures, and therefore the energy consumption of the motor, or the throughput and the yield.
In May, Augury acquired process-health AI specialist Seebo in a cash and stock deal worth an estimated $140 million. The target is the claimed $1 trillion in untapped manufacturing capacity from inefficient factories and processes.
If I detect a malfunction in a gearbox, and we know that the lead time for a spare part is eight weeks, we can tell the operator, ‘If you slow down the line, you can maintain throughput and quality, while increasing the lifetime of this component. But if you don’t do that, it’s going to fail and you're going to be shut down for a month.
We want to build a world where everyone can rely on the machines that matter, where the power is always on, and the water keeps flowing. But as we can see, that's not the world we live in yet. Every week you hear about supply chain disruptions, the chip shortage from China, the energy shortage via Russia and Ukraine…
What we do is at the most basic level: we work with the largest manufacturers, and we help them make their production lines more reliable, more productive, and more sustainable, so they can go and restock the shelves in your local supermarket.
An innovative and exciting venture. But can Augury’s data help with the big-picture problems we face: soaring energy bills and global warming, which may push many companies to the brink? Yoskovitz says yes:
Substantially so. Predictive maintenance can reduce energy consumption by up to 20%. And on the process health side, from Seebo, we can take data from existing datasets and help customers optimize their production lines to reduce energy consumption and their carbon footprints.