Earlier this year Plant Services and Augury conducted a research survey. Augury supports its customers by providing them with superior insights into the health and performance of their machines. The Machine Maintenance & Reliability study was designed to explore the common challenges to machine health faced by industry, and how organizations are putting it into action.
James Newman is Director of Market Strategy at Augury, with more than 20 years of experience focusing on understanding how assets and plants can operate better and more efficiently, and how that can be augmented by technologies. He recently spoke with Plant Services editor in chief Thomas Wilk on the universal challenges that industry is facing, and how data from the survey shows how those challenges might be solved together by those on the front line and in the C-suite.
PS: For this podcast, we're going to talk about a study that Plant Services and Augury did together. But before we jump to the results of the study, could you tell us a little bit about yourself and about some of the projects that you're working on right now?
JN: Sure. I've been spending more than 20 years now focusing on understanding how assets and plants can operate better and more efficiently. I started my career as an engineer, looking reactively at how plants operated and looking at things like trying to calculate remaining useful life, those kinds of activities. About 15 years ago, I was the victim of “complain too much and you’ll be asked to solve the problem,” so I began to look at how our processes could be improved by technology. Since then, I've been heavily focused in the digital transformation space, looking at how processes and the way people do things can be augmented by technologies.
The last several years, I have been very much focused on analytics and in particular in asset performance, and how we can drive that across a whole bunch of industries. At Augury, I am focused on market strategy, but that entails how we look at how our solutions can accelerate the ability to make assets run more effectively, to break some of the paradigms of how asset performance has been done in the past to get faster value. A lot of that's around how we bring new value and new opportunities to our customers. An example of that is our new release of a portfolio product called Supporting Equipment, which is all around taking super intelligent things we've learned on critical assets and expanding that to a whole new set of assets, to bring insights and let people manage things the way they didn't know how to do before.
PS: It's funny you mentioned how it works when you identify an issue and someone says, "Oh, how are you going to solve that problem?" That's how these career moves go. And I'm guessing that that's part of how this research project came about too. Could you tell us a little bit about the research initiative that you began? What prompted Augury and your team to do the study? And what kind of gaps in industry knowledge were you seeking to fill in?
JN: Augury talks to a lot of customers. We talk to people all over the world every day, and they give us a lot of feedback, which is really useful. But a lot of these studies in particular are an opportunity for us to make that universal. It allows us to look at, is this just the leading edge or the bleeding edge companies who are talking to us, who have these problems, or are they really more widespread? They allow us to begin to normalize some of these findings and these observations in a way that allow us to have deeper insights, but perhaps more importantly, to have a universal language to talk to customers: “These are your peers talking. It's not just us, not just Augury or a provider trying to sell you something. These are what your peers are saying. These are what universal challenges your industry is facing. Let's talk about how we solve those.”
So, for us, these studies serve two purposes. One, it gives us insight into how we can better talk to our customers; and two, it helps us really identify what the industry is saying, and so we can focus on how we can go and attack those major challenges.
PS: Well, for this study, the report that we released is called "Machine Health is Business Health." We're going to put the link to that report in the podcast notes for anyone listening. There was an infographic that was drawn from the report in the January issue of Plant Services, and there's also an article page we'll link to on our website with the infographic. James, for the survey itself, I thought it was really smart that the way it was structured. The survey forked fairly early on into two different audiences, one was called the corporate audience, and one was called the frontline audience. Could you talk a little bit about the kinds of knowledge you sought to gain from each of those sides of the business, and how they interact together?
JN: It's a really interesting dynamic between the corporate persona and the plant persona, whatever you want to call that second tier, the frontline, if you will, because they don't always see things the same way. The corporate folks are looking across a wide variety of assets. Typically, several facilities are looking at high-end across universal challenges, while obviously the plant people are focused on their day-to-day, what their plant has to do. So, it's a very different perspective, it's a view, a lens of the world about how they're tackling that. And so what these splitting it up like this allows us to do is get both pictures, what's happening when you look horizontally and what's happening when you're going very deep into your own vertical, and allows us to see is their commonalities? Are they seeing the same challenges? Are they seeing the same problems? Are they seeing different ones?
And if that's the case, if the second half is the case, which by the way is almost always somewhat true, it allows us to understand how to better approach solving both groups’ problems, as opposed to just focusing on one and missing the mark on the second. It also means for the organization, they can now understand are we actually aligned at the plant level and the corporate level on what's important? And if we are not, then perhaps that's something to go address before you talk about technology or anything else.
PS: Let's dive into some of the data points right away then. One of the key questions in the survey for both sets of respondents asked about their key operational challenges. And all respondents, whether corporate or frontline, cited supply chain disruptions as their top challenge. It's funny right now in Chicago, I can't find 3 ounce Dixie cups. That's the latest supply chain quirk; last year, it was baseball cards, right now it's Dixie cups. However, other top challenges included unplanned production downtime and workforce skills gaps or upskilling issues, and with both of those points, maintenance teams have a lot of immediate control over those. I'm curious, what are your thoughts on these data points on operational challenges?
JN: I think there’s a couple interesting points there. I do want to come back to supply chain because I think it's not as unrelated as people sometimes think it is. We start with unplanned downtime, workforce challenges. Really, we're talking about the challenge of, can technology taking the existing workforce and help them actually have an impact on unplanned downtime, right? Always an obstacle for many people. They view this challenge as “unplanned downtime is something we don't have a handle on. And, oh, by the way, can our people adopt these new technologies or not?” These people know how to do their jobs. I mean, been doing for 30-40 years. Can this new technology come online? Will they actually adopt it? So, when you look at that kind of scenario, it's not a surprise that those two things are very high in the list.
James Newman
What it also means is that solution providers today, and I'm going to pick on me for a second as a solution provider, we have the responsibility to solve them both in parallel. That's not so easy today because what we need to be able to do is take in and understand the mechanisms by which unplanned downtime occurs and be able to solve that challenge. And we can spend a lot of time talking about how that's different than it was 5 years ago, 10 years ago. It's always been hard for manufacturers who have legacy plants to put technologies in. It's a costly thing, particularly at scale. That's gotten better, but more importantly, it's what do you actually do with that information when you have it? So solving an unplanned downtime challenge five years ago is not the same as it is today.
But in parallel to that, the solution providers have the responsibility to make that not a data science project. Who is the end customer you're trying to make accessible to this information? Up-leveling workers shouldn't mean “send them all to become data scientists”, or “send them all to become reliability engineers” if that's not their job. But the solution providers now can take all that knowledge set and boil that down into something actionable, regardless of the skill level, that's easy to adopt and build into your workflow.
Let's be honest, everybody has a smartphone today. Well, my mom doesn't, but that's a different case. But most people have a smartphone, and they know how to use the apps on it. They don't require 25 hours of training to go buy a new app from their favorite app store and start using that. They do that and start using it right away. That's the challenge we have to solve in the workforce side. And if we can do that while solving unplanned downtime, that's the way we overcome those two challenges.
Let's flip back quickly to supply chains, because I want to make one point about that. Actually, that makes it more important to solve those unplanned downtime and workforce challenges, because when supply chain is constrained, unplanned downtime becomes even more important to overcome because in many of our industries, you only have a limited amount of material now. You can't get more. You certainly cannot afford for waste to happen. You have to eliminate waste, and unplanned downtime is often a cause of waste. So when you're under supply chain constrained, it actually becomes more important than it does when you have plenty of supply to eliminate these problems, because now you need the same people to be more productive and keep the plants running so you don't have extra waste, because you don't know when you're going to get more supply.
The supply chain disruption, often people think, "Oh, that's a separate topic," but actually, it's a driver to, in fact, improve the way you operate because if you can eliminate not the maintenance cost, that's not the problem here, but the waste side of that equation, then it becomes really useful to make those two things work in harmony to overcome the difficulty of supply chain disruptions, at least minimize a little bit.
Listen to the entire interview
PS: Right, I hear you. The storeroom and inventory issues are often at the nexus of what a lot of Plant Services listeners and readers experience when it comes to downtime, when it comes to planning out production. I want to move to a data point involving mechanical problems and the ability to visualize the condition of the asset too – that ties back to supply chain and that you don't know what parts you need unless you know what the health of the asset is, as you said.
The data points in the survey that turned up was that 69% of the frontline respondents, and this is your millwrights, this is your first-level supervisors, said the most common cause of unplanned downtime is mechanical problems. Yet 67% of respondents said they don't have the ability to visualize the real-time condition of their critical assets. That's only a 2% delta. It seems to overlap pretty neatly. What are your thoughts on that data point? Did that surprise you when you saw it? Is this your experience when you go into plants?
JN: So, unfortunately, it doesn't surprise me. It's very common, and it's not entirely unexpected. If you think about the majority of the facilities, regardless of where they are in the world, many of them have an installed base, a set of assets that didn't come intelligent. And many of them have not replaced those fully installed things. Until the last several years, it's been super cost-prohibitive to go back and "sensorize," I'm going to use that in quotation marks since you can't see my hands, "sensorize" those assets, and many of those are also invasive. If you go in and put in temperature meters, or flow meters, or other kind of things that are invasive to the thing, those have their own risks too. Historically in the past and wiring that stuff up and doing all the penetrations to get wiring cabling through the systems, those have all been super cost prohibitive. Nobody could do that at scale.
In the last several years now, we have this new set of sensor technology coming online that continues to improve and continues to be able to use things like W-Fi as Wi-Fi throughout the plant. Sensorization of those assets have become much easier than ever before, but that also takes some time. So, it's not really a surprise today that they say, "Oh, I can't really see that," because, in fact, it hasn't been accessible to them before.
That's changing, it's becoming accelerating. There are lots of people now, not just Augury, who sells sensors. That's becoming a place now where you can overcome that challenge. But let's also be honest, preventive maintenance is a safe way of doing business. Not necessarily the best way of doing business, but it's a safe way and a known way of doing business. And you don't need real-time condition to every three months change something, you just don't need that.
And so, we also saw in this survey, which I don't think we're going to talk about, but it's an important point that many of them still do preventive maintenance. That's still a predominant way of doing maintenance. And those two things go together. You don't need real-time assets if that's your main way of managing your assets. But if you want to do it better, if you want to solve unplanned downtime, if you want to solve production challenges, you can't stay there.
Now, for the first time, we're in an inflection point that, in fact, you can begin to do something about that. So, what I hope we'll see is when we redo this survey in two years, that number will have come down a lot, right? That's the goal. Now we're able to do it, let's now go do it, is where we are in the inflection point. And so what I hope we'll see in the next two years, four years, six years is the number of people who can't visualize it trending significantly down exponentially.
PS: That is interesting. I know that a lot of our readers are familiar with the idea of doing an asset criticality analysis to know which 20% of your assets drive 80% of the production, and yet that doesn't seem to matter when it comes to their ability to visualize the data, as you say.
JN: That's right. And again, because those assets are critical doesn't mean it was easy to go put sensors on them if they didn't come with them. And so, it is always been cost-prohibitive. We're overcoming that challenge now. It's now accessible at scale, and so that's where we'll begin to make a difference.
PS: There were some data in the report itself, and, again, the report's called "Machine Health is Business Health," that I was hoping we could talk about too. It's about access to failure data, especially about critical assets. More than 80% of respondents when they were asked about that data said, yes, they had information on why failures are occurring. It could be an FMEA, it could be root cause analysis. However, and this is frontline workers we're talking about, less than half said that they knew what the cost of the failures were to the business. And I'll be honest, that surprised me. I thought that more respondents would have an idea of what the value of their work was, i.e., the cost of a minute of unplanned downtime, that they would've an incentive to minimize that. Your thoughts on that?
JN: Unfortunately, it seems to be a recurring theme. Does it surprise me? Every single time it surprises me that people don't know this. But by now, I should stop being surprised about that because it's a repeated theme over and over again.
So, why is that? That's a really important question. I believe it comes back to how people are measured. And what I mean by that is when you look at how maintenance and...so, let me make my standard quip here. There's always been this ampersand wall between operations & maintenance. That wall is real, right? The maintenance teams, what they're measured on isn't necessarily the things that drive value to the bottom line of the business. Now, cost obviously does. But if you compare cost to revenue and production loss, those two things are not on the same scale.
And when you think about that, the fact is what we would get measured about is our cost of maintenance, or our cost of parts, our cost of overtime. Those things are super important, and we shouldn't discount them. But in the bottom-line analysis, when we're talking about things like sustainability, and production, and revenue growth, and those things, those two things are not equal, and they're not determinant. So knowing today, if you look at the measures, even things like OEE, which is the gold standard for managing how well you're managing your plant, don't care about profitability, really. They actually don't care about things like waste. If you waste more to have a higher availability, it's not in the metric. The metrics we're being measured to don't necessarily drive people to care about how much cost of downtime is.
So, ideally, we find a way to leverage information. I'm not going to talk about technology because that's really less important, but the information we can now gain to break down that wall and begin to have this idea that maintenance and operations are more tightly aligned, and as a result, we're being measured to the things that matter and not just the things we can measure.
It can also mean a little bit that we can begin to look and say, "We know a lot about our assets." People know how assets fail. We've known that for a long time. But now, we can say, "Why does it failing actually matter to the business?” It’s a new way of looking at that information, and that's a way that now we can begin, hopefully, to combine those things together, where process health and machine health come together to talk about operational health.
PS: Let's end this podcast by focusing on standardized work. You asked a couple of questions on that in the survey too. On one of the questions on the survey, frontline workers were asked how important it is to standardize a global maintenance reliability program across all their manufacturing sites. Of the respondents who chimed in on that question, only 54% said it was either very important or critical. Now, there's a lot of debates you can walk into a maintenance event with and start a fist fight in the room. This usually isn't one of them. Usually, people tend to agree that standardized work is key to success. What do you make of that 54%? Is that tale of the haves and the have nots? Is this one of the predictors that you would use to assess maintenance programs’ likelihood of success or failure?
JN: Yes, I think it's an important metric for success or failure, but I do think there's a couple of avenues to look at this from. So, if we take a look at it from the plant level, every plant is unique and that's true. Even if they were built to spec the first day, the second day, they no longer were exactly the same. It's just a fact of life. So, there are things that are different about each plant that have to be accounted for in maintenance and reliability strategy, and that's without a doubt the case. And so, every plant seems to think, "I need to run my plant the way my plant operates, I need to have data about my plant," and they're right. But what it doesn't mean, this is where often we get hung up in this topic, is there fundamentals and best practices and things like that we can use horizontally.
So, it's a horizontal versus a vertical thing many times, and we didn't get quite this detailed in the survey. It's a thing we could think about next time: is that view different between the corporate level personas, people who are looking horizontally versus the vertical people? It would be an interesting correlation to make. Historically, you see a lot of plants want to run the way they need to run their plant to get their things done. And corporate people are thinking about, "How do I get best across everybody?" And those views aren't the same.
What's also been true is that standardized work practices have been hard, right? So, APM, the Asset Performance Management topic isn't new, right? Twenty, 30 years ago, there were technology providers before I ever got into industry who sold APM solutions; intending to standardize all maintenance practices. But in fact, even today, all this time later, that's not universal and people don't always agree about that. So why is that? Because, in fact, the philosophy that's used is, “go build a bunch of models and then go see if you have the data to support the models that build out your maintenance strategy”. And that's not always so easy and every plant doesn't do it the same way. That's a difficult thing.
What's possible now is the fact that we can now flip that on its head. Instead of creating a strategy and then going and find the data to support the strategy, now we can begin to look and say, "What is the data telling us about the right way to manage assets at the plant level? And what can we learn from that in order to learn across the corporate strategy?"
The corporate strategy can now be informed by the data that's coming off the assets that is driving individual maintenance strategies. And the fact that becomes a new harmony of the way people can look at it. Before, in fact, they can't. And so, I think that's a really important point where we are now that we can now use data-driven decisions.
What it also means, though, is as we get there, now we can begin to look horizontally and leverage these standardized practices to begin to help understand why do the best guys perform the best, why do the worst guys perform the way they do, and normalize some of the uniqueness to the standard so we can now say, "Okay, how do I bring this guy up while sustaining this guy at his high level?" And it becomes a new way of looking at it. But it’s at the right level. It's understanding how you take and normalize the right things and leave the things that are specific to a particular plant available to them specifically.