Scaling Success: The Culture Shift Required For Predictive Maintenance and Digital Transformation
Show notes
Welcome to the Trend Detection podcast, brought to you by Senseye Predictive Maintenance – which gives you visibility and insights into all your assets, from single machines to full plants to help you reduce downtime, increase knowledge sharing and accelerate digital transformation across your organization.
You can find out more about how Senseye Predictive Maintenance can reduce unplanned downtime and contribute towards improved sustainability within your manufacturing plants, by visiting: www.siemens.com/senseye-predictive-maintenance
Transcript
Welcome to the trend detection podcast powered by Siemens, and now your host, Nel Sullivan. Hi everyone, um, I'm Liz McGinn and I'm the host for our panel discussions today. So today we'll be talking about scaling success, the cultural shift for predictive maintenance and digital transformation. I'm joined by three panelists today, Nick, who you heard speak earlier, Natalie, who you heard mentioned during the BlueScope, and then Todd, um, who's recently joined us at Sunsai. So, could you please introduce yourselves before you get started? Nick, could you start, please? Hi Liz, um, yeah, so, uh, I think everybody heard the introduction earlier on actually, so, uh, for those who missed it, um, I've spent 30 years at the intersection, uh, of transbusiness and technology, helping companies, uh, improve their operating performance through those, those tools. Awesome, thank you, Natalie. Hi, everyone. Um, I've been working with Sensai for the last, I think, 4.5 years now, uh, and the delivery of the tools. So I worked with BlueScope and other customers to get them on boarded and trained up and make sure that Sensai is doing what they want to do. Awesome, and Todd. Uh, thanks Liz. Uh, listen, I'm Todd Martin and I am the Siemens Sensi managed Services lead and I recently joined the Siemens team after retiring from, uh, Nissan, um, where I used, uh, Sensai for 7 years, um, was involved in the. Um, the proof of concept that brought the um system on board there and, you know, implemented, designed many of the data collection systems to provide the data to the Sensi system. Awesome. Thank you. So, let's jump in. Um, before we can talk about what needs to shift in a maintenance team or a company's culture to implement predictive maintenance or any new technology, we should first kind of figure out what's currently happening within the maintenance team. So, Todd, as you've worked, uh, you know, in maintenance before, what's currently happening that you're, you saw during your experience? So especially in a large facility, you know, your maintenance teams are broken up into a lot of different departments or groups that, you know, focus on certain areas on there so a lot of times they may have their own individual ways of doing things, um, so that's something that you have to consider, um, each of those groups probably have different KPIs based on the processes that are going on in those areas, um, so. You know, also their training may be different, um, just because it's different equipment. Um, So there's a lot of different experience, backgrounds and skill sets that are um going on in those um different groups. Yeah, the training is uh interesting. It's kind of callback to BlueScope talking about their training. Natalie, do you, um, want to kind of speak to that? Were you involved in their training aspect of it? Yeah, Bluescope was a really interesting one, So for, for training them, I worked hands on with them and just getting them in using the app and figuring out how to monitor their equipment. So we had lots of one on one sessions with myself and Chris Watson to figure out how these machines work. They did not have traditional condition indicators. They did not have vibration monitoring on the rotating equipment for the most part. So we had to get really creative. And throughout that process of figuring out what to bring into Sensei and how to configure it, they really learned a lot about how the app works. Nice, awesome. And they mentioned also during that time that they talked about this role of like figuring out who can be champions and the importance of having that. Yeah, so BlueScope, they, they really did a lot of things right, and I think one of the biggest things that they did that led to their success is that they had drive from the bottom up, the end users in all the different areas, their site, their sites are massive, and they wanted to use it. But they also had support from the top bottom to, uh, incentivize and engage their end users, uh, into using Sensei and to really integrate it into their, their system. So, I think that was their, a really key part of why they were successful. And actually when I went on site to do some introductions and to do some training, we had one session planned in which people from the site could come in and sit in and learn about Senseii and figure out how they can get started. And there were so many people that showed up that we needed to make a 2nd 1, so that everybody could join. So I think, I think the key in finding the right champion is whoever is sponsoring the project needs to sell Sensai internally so that people can see the benefits of it. People should want this because it will make them more efficient in the long run. Yeah, Nick, you mentioned something similar during your presentation is kind of what does this mean to me versus just what does it mean to the company? Absolutely, and, and, as Natalie was saying a second ago, those champions, I sometimes I describe them to to companies I work with as Yodas. Yeah, it's, who are the Yodas in, in, in the production facility that these are typically the people that have been there for 30 years that when something goes wrong, they know how to fix it. It's not necessarily written down somewhere, they're just the wise people. If you can win those people over. That's so infectious in terms of getting everybody on, on board. Uh, and I think the, the other thing that I think is linked to the training is, I, I've seen a lot pilots and projects get deliver the training to the end user. And that that's great, but if they don't keep using the tool. They forget how to use it, they forget about the benefits, and it becomes another one of those um implementations that's gathering dust on the shelf. So it, I, I think in addition to just that, that training motion, it's keep that continuous engagement. You know, I talked about some of that feedback loops, yeah, what's working, equally, what's important, what's not working, and keep that improvement, making things better all the time. That's awesome. I mean, you mentioned, I guess, this whole, um, feedback in the training loops and the training, sorry, training and also this champion. I know Todd has some experience kind of, you know, pushing the sense I kind of predictive maintenance digitalization journey forward. Do you want to speak to a little bit um on that? Yeah, sure. So yeah, definitely you've gotta have champions. One of the key things you, you, you want your champions, you wanna pick them from a pool of people that are excited about the technology or technology in general, um, you know, you're gonna have different, um, age groups within the maintenance departments you're gonna have your, your newer techs, then you're gonna have your older techs, you know, you'll get your most resistance from the older techs because they're old school. So a lot of times it helps you grab those younger guys who have grown up with the technology um and that's really where they're focused at on this new cool, you know, software application um and are hungry to make a name for themselves. You know, in their group, you know, to stand out, so it's kind of important that those are really good resource, you know, ones to have as a champion, um. Um, So And and the other thing is too is not to have too many of them. Um, you, you really, you really wanna have potentially one oversight on here to make sure that it's being utilized, you know, properly. As we, as I saw in the past, you know, some areas are gonna have more benefit from, um, from Sensei than other ones, it's just the nature of the type of pieces of equipment. Some wear out slower than others and have problems. Some are faster, the faster ones you see more benefit from. And you know, so it's, it's keeping them engaged and. Across the board I think Todd, just to build off a little bit what Todd was saying as well though. I think it's really important to think about. These tools like Sensai and predictive maintenance, they're not just the tools for the maintenance teams, you know, I, I talked about creating more visibility and curiosity around this. So if you start showing the machine health to the operators, not the maintenance teams as well, that becomes part of their daily routine. It, it becomes something that they care about, not just the production run. So it's about elevating that same information and making it more broadly accessible and visible also helps that journey significantly. Yeah, you mentioned a really nice phrase there that it's a tool to help this predictive maintenance journey. I know, uh Natalie loves to use that phrase. Yeah, I feel very strongly about that. I think there's some common misconceptions that Sensai itself is predictive maintenance, but it's not. Predictive maintenance is a whole process and a whole strategy. Senseii is a tool to help you achieve that, but you have to use it, and you probably have to change some other practices as well in order to actually achieve predictive maintenance. So let's say Sensai tells you about something, let's say you have potentially 4 months before you need to maintain a piece of equipment. What are you going to do with that information? So, That's something that we, we often help our customers with is like, OK, you got all your data in, but that's just one small part of the puzzle. What are you going to do with it? Yes, definitely. Oh, go ahead. No, I, I wanna say I agree 100%. It's just like, um, you've got to take that information and you might have to change your um maintenance practices to, um, to handle that. You're gonna learn probably way sooner that something's going on with your equipment than what you can justify actually doing the work on it yet. Or you might like Mixer, you, you might work with operations to change how you're using the equipment to make it last a little bit longer to, to baby it a little bit so it makes it to the next shutdown, things like that. How is that affecting um kind of the KPIs that people are looking at when they're, you know, investing in this kind of thing. So the KPIs can be really difficult to calculate, especially when there's, there's so many unknowns. If you're doing predictive maintenance, the thing hasn't failed, so. It's hard to anticipate and like in every case, what might have happened if you hadn't done it, um, but it is something we do work, it's hard, but we work closely with our customers to try and put hard numbers to these things. Uh, a lot of times people come up with kind of generic KPIs and so we, we try to make them quantifiable. We, we na, I mean, we talked earlier on about the disconnect between operational KPIs and the financial metrics that the business runs, run against, and I think. For most businesses I work with, that that that that connection has to be strengthened over time, because an OEE, a 20% OEE improvement in OEE or a 20%, uh, 15% reduction in unplanned downtime, what does that really mean and who really cares about it? But also, I, I think we should also think about KPIs in terms of making them not just financial, but relevant to the people that need to make decisions. And that doesn't necessarily mean managers. That could be the maintenance teams, that could be the operator. It could just be a health statement or a health score of my machine, and we wanna keep it green. And we wanna keep it green 90% of the time. So you don't, you know, it, they, they all roll up, but it's actually, you know, I talk about buzzwords, yeah? It's, it's making things simple. How many, how many Big Macs did we not, we saved this week, yeah, or this month? It sounds stupid, it may be slightly, it might be slightly, but it kind of makes it relevant. And I think that's the key for KPIs rather than just pure financial statements. Yeah, it makes sense to have this kind of tangible um thing cause, you know, people can't really imagine how much is actually being saved, but if you put it in a tangible, it makes a lot of sense. You could, you could bonus on, you could bonus on big backs, yeah. Exactly. And I guess the other thing that we uh kind of spoke about is kind of the time, another tangible KPI that is the time that you'd spend fixing things because you can kind of quantify how long it is fixing these things. Mhm. You fix it, you can fix it during a breakdown. Or um you can fix it, do, you know, scheduled downtime, avoiding overtime. Yeah. But, but I, I, I think Todd, you were using a French phrase earlier on called wrench time. How much wrench time is a maintenance person? Actually got fixing things rather than analyzing data, and I think there's a combination of what I just mentioned and making SIT KPIs really simple, so it gives direction to the people to do something, and then then they're not, not, not spending time analyzing why they should be doing it, they know what they need to do, and they go off and have the maximum wrench time poss possible. Yep. You know, wrench time repairing or um doing preventative maintenance. Nice. It's a really great point and I guess um I want to circle back to kind of this point of our whole event here is the scaling. Can you guys speak a little bit to the importance of scaling? Uh, I can get started. Sorry, Todd. I'll go for it. Uh, so Sensii really shines at scale, and I think one of the beautiful things about Sensai is if I have, say 10 motors that are roughly the same, same data coming in, or if I have 100 or even 1000 motors coming in. It's pretty much the same amount of effort to configure 10 as it is to configure 100, the way the app is set up. So why not get 100 or 1000 in and give yourself more chance to to catch something, um, if it's all the same to, to you, to get the data in there. Um, and then another part of Sensai that that is often overlooked if you're not on boarding many assets at the same time, is the capacity to view your assets in a fleet. So if you have a whole line of assets and they're performing roughly the same thing in the same way, you can look at their data together and see outliers that they may have been on board in a stable state, but they may be sitting, you know, if everybody else is sitting here, one might be sitting up here, might be stable, but very different, and you might want to know about that. Um, so some things like that, I know BlueScope use that view a lot in the app, and if you don't scale and you don't have that duplication of asset types, then you don't really get to see it. Yeah, definitely, um, you know, scaling, you know what helps is is that you're gonna go into this predictive maintenance journey on here, you know. One of the easiest thing to do is, OK, we're gonna pick these few assets and we're gonna, we're gonna start monitoring them all that, but. Of course it's gonna be successful, but how do you make it successful on everything else and as easy as possible, so going into the journey, thinking very high level outside of the box, um, determining. How do I set this up so that. I can easily bring more and different types of assets on there, you know, do you have existing um condition based monitoring systems that you're gonna use, you know, so how do we access those and make it as easy as possible on there or if I'm gonna bring in new one, new condition based monitoring systems, let's do it in such a way that I can just, you know, use all these pieces, the same pieces over and over again, maybe you do use them in a very creative way. To get that information and it just it's easier and faster to implement the system. Yeah, I think definitely when you look at that, Todd, as you mentioned, that repeatability. That's one of the true paths, but that, that almost, as I mentioned earlier on, had to, you have to start before you're in the implementation. You need to do a bit of work understanding how many of those assets are similar, how, what are the things I wanna measure, are they consistently measured in the same way? Don't get caught out with different time stamps between different facilities in different locations, stupid things like that. I've seen get catch companies out of time. And time again, so they lose the relevance of it. Um, I was really interested in what Natalie was saying as well about the, being able to classify or group assets together. Um, you know, one of the things I talk about a lot is about the true potential of, of, of manufacturing. So it's not about how, where you're performing, it's actually the difference between your average performance and your lowest uh worst performer. And understanding that difference and why that might be, and if you can bring that up, and the whole, your whole system improves. So that true potential view, that that's a really interesting way and a nice actually KPI you can then look at and look at the actual knock-on effect in terms of financial performance then. And I think when, when you go through and you look at KPIs and you figure out what you want to monitor and and what you want to track, it will lead you to the conclusion that it's best to scale. Sometimes folks just want to come in and put their most critical, most problematic assets into Sensai, but really what are you going to get out of that? Um, may, it may give you something you didn't already know, but you're probably already over monitoring that asset. Why don't you get in full lines of asset and reduce your planned maintenance and, you know, track things that maybe you haven't tracked before. There's a lot of possibilities when you're on board at scale. Awesome. So we're about 20 minutes in, so we've got about 10 minutes left and I'd like to um kind of jump into some questions. So please, if anyone has any questions for our experts here, um, throw it into the chat. Um, so for first question, curious, uh, Here's customers want to start a PDM project with a POC, but from my point of view, since that PDM projects don't include a POC phase in the process models. Is this right? How should I encounter um the customers in this matter? So, one thing that we are trying to do is kind of change the language a little bit, cause you're right, we, we don't do it necessarily proof of concept all the time, um, because Well, it works, it's going to work, right? Um, I don't know, I can say that. Um, what, what we're calling it instead is an initial deployment. So maybe you do an initial deployment of 50 assets, narrow it down to just a handful of asset types. So 50 assets, maybe 3 different types, motors, fans, pumps, something like that. And that will still help you see value because you have repeatability in there, um, and just try to, you know, make sure you're selecting assets that actually have a chance of giving value in the timeline that you're looking at. Hopefully, that answers that question. Awesome. I, I, I think, I think we also just need to recognize that as a manufacturer, and I, I, I, I, I have to say neutral in this, yeah. There's skepticism, there's money on the table here. They want the proof is always in the pudding, as I, I don't know if that translates, it goes across the pond to America well, but in, in we said the proof's in the pudding. So, yeah, a POC is always perceived to be a lower risk approach to go in to demonstrate the value. What I see going wrong is there is no plan to scale thereafter. So We should always go into a proof of concept with if we can demonstrate success, this is the plan on day 2. Or, or actually day one, after the end of the POC where we start to deploy more broadly, and, and that's when we talk about scale, but that's when the value increases for the customer. They've proven it, they're comfortable with it, the skepticism, yeah, and we've always done it this way, has gone out the window because they've seen it as a new way, a better way. And they moved forward …