Predictive maintenance: It's a way of life not a technology - with Richard Jeffers
Show notes
Welcome to the Trend Detection podcast, brought to you by Senseye Predictive Maintenance – the platform which enables predictive maintenance at scale across all of your assets, across all of your plants.
In this episode, we discuss:
From Breweries to Maintenance Strategy: Richard Jeffers shares his fascinating career journey from running and maintaining breweries to leading technical and solutions teams—highlighting how military leadership shaped his approach to asset management.
Why Maintenance Needs a Seat at the Strategy Table: Richard explains why maintenance professionals must understand their company's commercial strategy to align maintenance decisions with business goals and manage risk effectively.
Technology is Cheaper—But Strategy Still Wins: While predictive maintenance and IoT tools are more accessible than ever, Richard stresses that the fundamentals of a good maintenance strategy haven’t changed. Technology is a tool, not a strategy.
Balanced Maintenance, Not Buzzwords: He emphasizes that a successful approach blends multiple techniques—including run-to-fail and predictive methods—based on business context and asset criticality.
Leadership Drives Culture Change: A clear maintenance vision, strong leadership, and cross-functional collaboration are critical to embedding effective maintenance strategies and embracing new technologies across an organization.
Transcript
You know, the ones that are more closely aligned to the corporate strategy, those are the critical initiatives, um, and, and the ones that, you know, there'll be, you know, other stuff happening, but if it's not, you know, laser aligned with the, er, with the corporate strategy, then you should be putting less resource into it. Welcome to the trend detection podcast powered by Siemens, and now your host, Nel Sullivan. Yeah. OK, so we're on. OK, cool. So I'll um I'll just pause and then we'll go into it. Hello and welcome to the latest episode of the trend detection podcast. Um, I'm really pleased to be joined by one of our, I'm gonna say regular guests because you have been on before. That means you are a regular guest now, Richard. Um, so yeah, Richard Jeffers is joining us. Um, we had a really great episode in the past, but today we're actually talking about predicting maintenance. Maybe not too much of a surprise. But actually, you know, the topic's actually, it's a way of life and not just the technology, which, um, I think you did a presentation on this recently, Richard. So we're gonna just sort of dive into that, the content of that, um, but it's something we've talked about a lot within Siemens as well. So the emphasis on, not the emphasis shouldn't be, sorry, just on the technology change, but actually the, the people change and the cultural change. Let's say that's required, but, um, before we dive into it, Richard, it might be good just to remind people or new listeners, um, who you are, what you're doing at the moment, all that kind of thing. Uh, excellent. Well, thanks for, uh, having me back, Niall. It's always, uh, always good to talk to the, to the gang at, uh, at Siemens. Uh, I think this is my 3rd time now, um. But, uh, but my first time as a, as an independent, uh, and, uh, you know, for the last 4 months I've been running my own small asset management consultancy, focused on, uh, Cape, uh, maintenance strategy and industrial data. Excellent, excellent, and welcome back. So I must have, um, I've interviewed so many people, I forget how many times people have been on, but that's a good reminder. Thank you very much for that. Um, so yeah, it's let's look, open the topic up straight away. So what do we mean by this, this concept of predictive maintenance being a way of life and not a technology? So yeah, just to start the conversation. Uh, you know, I think, you know, the world has known for at least 40 years, the principles of predictive maintenance. You know, since, since Nolan and Heap did, did their seminal work in the, uh, you know, the, the aviation sector, we, we've understood the concept of failure curves. We've understood the concept of the P2F interval, uh, and the importance of, of collecting leading indicators of failure and responding to those. But, but actually, despite having known this for 40 odd years, you know, many, many businesses are still, you know, striving towards an aspiration of predictive maintenance. Uh, and for me, that's because of, you know, cultural, uh, or, or, or, you know, challenges, or, you know, cultural ways of working challenges they've got in their business. They're not being data-led organizations. So, so whilst, er, you know, both the, you know, the Sensi PDM platform and, and others are fantastic at processing that data, er, you know, if you don't have that culture of responding to data, er, then actually it does just become, um, technology, rather than, er, something creating value. And, and so I guess you sort of touched on, um, the question I was just about to ask, but, um, I guess, you know, in a wider digital transformation program that, you know, there's lots of different technologies and things like that. So, um, I guess the question is why, why do people see PDM as sort of, as, you know, as a software tool and not this sort of, you know, wider initiative, let's say. Uh, uh, you know, I think, you know, people are. You know, struggling, a lot of organizations are struggling to build a coherent maintenance strategy and understanding all the different things that need to go together to, to drive that improvement in, in Up time. You know, it, it is about, um, you know, predictive maintenance and condition monitoring, but it's also about work planning, it's also about storing planning, er, you know, it's about the ability to write really high quality work orders in your CMS. There's lots and lots of things that have got to happen. Simultaneously, to, to move yourself up the, um, the maintenance maturity journey. You know, there, there isn't a shortcut that says you can, you know, technology where you're out of that problem. You know, if you don't have an efficiently run storeroom, you can respond really quickly to the alerts that a condition monitoring's, uh, creating. But you're still gonna have a really long time, long mean time to repair, because that spare part that you need is not available. And that's just one example, er, of, of all the bits that need to come together at the same time, er, rather than just, um, you know, 11 slice of that challenge. So, so I guess what we're saying here, there needs to be a maintenance strategy, but I mean, it maybe it sounds obvious, you know, saying it out loud, but maybe it's not so obvious to, to people on the ground. We're busy, um, you know, busy day to day lives, but you really need a, a maintenance strategy to lay the groundwork for techno technologies like predictive maintenance and others. Absolutely, and, and not just a maintenance strategy, but a maintenance strategy that's keyed into your corporate strategy. You know, I, I, I talked to, to a lot of maintenance engineers who, who go, you know, how do we persuade the board, uh, you know, to invest in maintenance, or how do we persuade the exec to invest in maintenance? Uh, and, and, you know, one of my questions is, when was the last time you read the annual report? Do you understand what your business is trying to achieve, what the corporate goals are? Because maintenance is not an end in itself. You know, it's an enabler to deliver your corporate, commercial, and, and brand strategy. So, really, like, like everything else in maintenance, it's gotta start with that strategic plan, er, and then cascade its way down through all the different elements. And then, and then of course, you know, investment in condition monitoring and predictive maintenance should play, er, you know, an important part in that strategic plan. And, and I guess, yeah, uh, I mean, we talk, we talk about this, you know, this a lot, this, you know, connection between, you know, departments you mentioned, sort of, you know, the corporate strategy down to the maintenance team, but it's sort of a cross organization as well. And you mentioned also, um, sort of earlier, um, you know, at the beginning of this episode about, you know, culture as well. So how does, how does sort of culture impact the success or failure of predicting maintenance and And that, and maybe you could refer to that, that example, you're giving that sort of disconnect between the corporate strategy and what you're, what people are doing, you know, on the shop floor, for example. Yeah, I think so, so there's probably, you know, there's a couple of things there. I mean, you know, the disconnect on culture, you know, if you've got a culture where you're rewarding the heroes for responding to, to firefighting, er, and, and meanwhile, you know, ignoring the, you know, the people in the back room who are busy building data models and respond to that data. Then, then actually people are never gonna, you know, value the work that's done preventing failure. Um, and, er, you know, I did some work in a, in a, in a food factory, er, a few years back, uh, and they were absolutely trapped into a reactive cycle. You know, all the cultural triggers, all the, you know, emotional reward mechanisms were all focused towards the, the heroes, in inverted commas, who, who, you know. You know, turned up at 2 in the morning to get the, uh, get the factory back up and running again. You know, meanwhile, there was a young lady, uh, you know, in the corner who was busy, you know, looking at data and going, we need to do something now to stop this failure. But actually everybody was too busy running around fighting fires, uh, to go and, you know, respond to the data that this, uh, this young lady was generating that would have stopped those fires light in the first place. Um, and, and, you know, in a lot of ways that, that business just didn't see maintenance as part of the corporate strategy. It felt maintenance existed, you know, separate to everything else. And, you know, I, I hear a lot about, you know, the business won't invest in maintenance, you know, the business won't give us the, the money we need. Um, but I, but I feel that so often maintenance professionals don't communicate the risk and benefit of investment in maintenance. You know, maintenance is an insurance policy. Um, you know, and, and like all other insurance policies, it's a choice. So, you know, you can make a choice to, to run to fail, or you can just run to fail. You know, one of those is a strategy, and one of those is, is, is a failure. Um, so if you've chosen, based on my corporate strategy, based on my criticality analysis, that, that run to fail is the right approach for this asset, fantastic, absolutely the right thing to do. Uh, what, what's not acceptable is if you've not made that choice because you've not understood the importance your assets play in delivering the corporate strategy. Yeah, so I think what, what we're saying is there's, there's often sort of a disconnect between, between the two. There's that this big, big, maybe like information void or understanding between the two, you know, two very, you know, very different, um, sets of stakeholders as well. Yeah. Um, but, but I'm really, I'm really interested by that, that idea that, you know, you know, firefighting is seen as, you know, success, I guess, because I could always almost imagine in, in my head as a way of technology solving issues. So it's so quiet. But not say everything's solved by technology, but it's a lot quieter, so it's sort of, there's almost suspicion like, you know. You know, because everything's running so smoothly maybe with additional technology that there's almost a bit of suspicion around technology if you've got that sort of mindset. Um, I, I, I, I think, you know, I, I, I, I've probably not seen that Niall, that, that suspicion around technology. I, I've just seen the, you know, the challenge of, of getting businesses to create the headspace for structured, uh, improvements in, in maintenance as opposed to, to reacting to it. And it's really easy to see, you know, activity being done by people responding to failures. You know, it's much, much, much harder to see the activity that's done by the people who are ensuring those failures never happen in the first place. You know, the, the, the, you know, the, the team that invest in making sure the work orders are correctly written, so that, you know, the, the, the right, you know, the right inspection standards are there. That they've got a, a really robust condition monitoring strategy, so they know which assets to go and look at, er, and they've got a really well run spare parts organization, so that the required spare parts are always there when they need them, er, but not when they don't need them. You know, that, that's all very calm, quiet, and, and actually, you know, under the radar stuff, um, and, and it's not, you know, in a lot of organizations, it's just not seen and, and rewarded. Um, because, you know, they, the, that activity of being reactive is, of course, highly visible. Yeah, exactly, because you can say, oh, I saved this major breakdown. It's like big, big headline news in, in your face rather than the, you know, the quiet work going on in the background where, you know, it's like we're pointing you to this issue which I fix proactively, doesn't quite have the same shine to it, but, you know, obviously has, uh, you know, an amazing impact. So is it a case, I guess as well of, you know, being able to sell these successes better internally or teams need to be equipped to be able to sell. Yeah, yeah, I'm, I'm, I'm, I'm gonna go back to that alignment with the corporate strategy, Niall, because for me, this is, this is absolutely fundamental, and, and if you look at the principles of ISO 55,000 and the, and the concept of the strategic Asset Management Plan, you know, that, that's really clear that maintenance exists within a corporate context. Um, so if you understand that actually, we've got to deliver, you know, this thing, uh, you know, and that requires an availability of, of X, Y, Z. Um, then it's down to the maintenance function to define how they achieve that availability, uh, and that's clearly our job as maintenance professionals. But, but actually, we can't define what, what availa, what's the right availability, uh, for, for our organizations. You know, that needs to play its part in a, in a, in a broader strategy. Um, you know, my, my, you know, in a previous life, uh, you know, I worked for Heineken, um. And, you know, breweries were there to deliver, you know, a brand strategy and make sure that we had, you know, stock available on the shelves at, at the right points in time. If you had a failure, uh, of your packaging, you know, in the run up to Christmas, uh, you know, that, that was big news, and we, we saw it with Diageo last year with the, you know, breweries running, you know, pubs running out of Guinness, um. But you know, what, what would be great to unpack there was what were the, what was the mean time between failures and the mean time to repair, that, that was sitting below, you know, some of those numbers. Um, and, and actually that's a really good example of where, you know, actually, if you want to achieve, you know, 1% availability of your products on shelves during peak, peak trading. You know, you can cover that partly by stock, er, but that's quite an expensive way to cover it, or you can cover it through high availability and, and high repairability, um, which is clearly a cheaper way to achieve that same, you know, product availability and shelf. Um, so you can imagine now, in the, in the, in the Diageo business, people are looking at, you know, why did we run out of product during, during that peak trading. And what are the actions we need to do differently next year, er, in order to ensure that we don't see that, that repeat itself. And I, yeah, it's interesting, so I guess if you talk about a reactive strategy, you know, you know, you can, you can, we said we can, you can see the success quite clearly, you, you know, you make the repair things back up and running, but I guess what we're also saying is, yes, you've repaired it, but that you can't investigate, you know, why it happens, where you can improve next time, the underlying data, you know, reasons underneath that, right? Yeah, and, and that's back to the cultural, you know, issues, uh, again, because, you know, if you've, you know, Structured problem solving, uh, you know, is, is again, it's a culture thing, you know, the, the desire to prioritize structured problem solving. Again, if you sit in a highly reactive organization, you know, you go from one fire to another fire, and you never stop to go, why, why is my factory bursting into flames, you know, all the time, er, and actually that, that, that structured problem solving at the end of each failure. You know, it is, is massively important because, you know, those failures won't have just happened, you know, we all know that, you know, only 17% of failures are time-related, the other 83% of failures are, are caused by, you know, non-time related failures. So all of those non-time related failures, you know, there, there can be, you know, a. Uh, you know, conditional, you know, conditional inspection, um, that will help you anticipate and therefore, er, mitigate the impact of that failure. Um, but, but you need the culture to want to build those, um, er, and once you've got the culture, then, yes, you can, you can have the software tools to help you do that at, at, at scale. No, that makes, that makes a lot of sense. Um, but, I mean, even if we consider, you know, that, um, or, I mean, there are obviously many organizations, you know, ready, have the mindset, you know, for using data-led approaches and predicting maintenance to, um, to do that. But then when, when they look at the market, which obviously Sensi is a part of, but look at the market as a whole, we know that there's lots of different providers saying they do predictive maintenance or they do AI or they do, and is it condition monitoring? Is it predictive maintenance? And it's a very confused, you know, market. We're in, and which I'm guessing leads also to a lot of sort of myths and misunderstandings about what predicts maintenances and how it can impact. So surely that, that probably contributes into this as well, but when they think they're getting one thing and it turns out, I know that's something completely irrelevant to what we're trying to achieve. I absolutely, Niall, er, there, there's no doubt that there is a lot of confusion, you know, out there in the marketplace, uh, around, uh, you know, what you, what you can and can't do with data, and indeed how you can and cannot collect data. Um, and, you know, you, you see different approaches to solving this problem, you know. You know, every trade show I go to, I, I will see the, the sensor led solutions to predictive maintenance, where you've got a, a single sensor that you mount on a, you know, on a, on a single asset, and it sends it to a, a specific cloud platform to look at, to look at that data set. And, and those, you know, those, those, you know, they have a part to play in my mind, er, in, in some applications. But in a lot of ways, you know, people are deploying those because they're not joining up cross-functionally. Uh, Around to what role industrial data plays in the corporate strategy, I'm gonna go back to that strategy piece again. Um, because actually, if you look at this from a, a strategic perspective, you'd actually say, in order to deliver my brand strategy, I need to drive availability. You know, that, that's not just a job of the maintenance function, it's a job of, of, of operations, but it's also a job, uh, you know, of the IT function as well. Um, but actually you see in so many places, there's a disconnect between the, the operational technology guys sitting inside the factory wall, uh, and the IT guys sitting outside, uh, the factory wall. And, and therefore, you never actually develop an integrated data platform, an integrated data approach that allows you to take all the relevant information you need from all the different systems. Uh, and, and, and feed that into the appropriate platform. You know, and, and what happens is the operational guys, the, the, the on-site guys use, uh, you know, shortcuts and stand-alone systems that, that don't sit as part of an overarching, um, data strategy. So not only does your entire asset management approach need to sit as part of a corporate strategy, within that asset management approach, there should be a data strategy that sits in there as well. You know, you'll have got a spare part management strategy, you've got a condition monitoring strategy, er, you might have got a lubrication strategy, etc. but data needs to be a strategic. Decision as well, and, and bringing together the, the IT view of data and the OT of data, uh A202, the OT view of data as well. And once you bring those into a strategy, then it allows you to say, right, now let's look at the right data management approach. …