Trend Detection Revisited - Bluescope Steel Success Case - with Chris Wonson

Trend Detection Podcast ·

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.

This week, Trend Detection revisits a standout episode from the archive.

In this conversation, Chris Wonson shares the story behind the deployment of Senseye Predictive Maintenance at BlueScope Steel and explores how predictive maintenance was being applied across its operations.

The episode covers:

How Senseye Predictive Maintenance was deployed at BlueScope Steel

A success case that helped avoid 24 hours of unplanned downtime

How Senseye Copilot was supporting the way teams worked at the time of recording

Practical advice for manufacturers implementing predictive maintenance

This archive episode provides a valuable customer perspective on scaling predictive maintenance and turning asset data into practical maintenance action.

Find out more about how Senseye Predictive Maintenance can help manufacturers reduce unplanned downtime and improve maintenance efficiency across their plants by visiting: www.siemens.com/senseye-predictive-maintenance

Please note that this conversation was originally recorded and published previously. Some product details may have evolved since the original recording.

Transcript

Welcome to the trend detection podcast powered by Siemens, and now your host, Nel Sullivan. Hello, and welcome to this episode of the trend detection, um, podcast. Um, I'm really pleased, it's, uh, quite a special episode today. And we've got, um, one of our customers, Chris from Blue Scope Steel. Um, and we've got a really interesting discussion today about, it's actually about a particular success case, um, that, um, that recently came, came through to us, which is really fantastic. We're gonna dive into exactly. Um, what, what had happened and how that was resolved with the help of, um, of Sensai as well, but I guess before we get started I'll just hand over to, um, to Chris just to introduce yourself. So over to you, Chris. Yeah, good day. My name's Chris. I work for BlueScope Steel out of our Port Kembla site, which is in Australia. Um, I have a background in maintenance. I've always worked in maintenance ever since I left school. Um, I joined BlueScope 1718 years ago, and I've worked at the Metal Co lines for most of that time. Uh, start of 2022, we started a pilot program in my native department, and that was to introduce predictive maintenance. We chose Sensei over three or four different platforms, um, and that was mainly around the, the AI component and the ability to, to tune it. Sort of automatically with that thumbs up and thumbs down feature, that was the initial attraction. Things have changed remarkably since then, um, we've had 3 or 4 massive sort of step changes in the capability of the app, and, you know, we went through the pilot program, it took roughly 12 months, we onboarded around 300 assets across 3 metal coating lines, and yeah, once the pilot was complete, Uh, we decided to roll out, uh, Australia wide, basically, um, and that's what I'm responsible for, so I'm predictive maintenance lead for Bluescope Steel, Australian steel products. We do support our global roll-out as well, which is pushing into US, China, Thailand, New Zealand, and a few other bits and pieces around the place, so, I have the awesome job of, of leading that, um, that regime, I guess you'd call it, so we've been going since 2022, and yeah, we're, we're just about to, to kick off in the US we're, um, probably 6 months deep in, New Zealand, maybe 18 months deep in Western Port or Melbourne. And yeah, and across here at Port Kembla, we're pushing into most of our business units. So it is, it, it's really cool to be part of, and, um, you know, our team's growing, so we're trying to accelerate how quickly we onboard and, and get things, uh, monitored through Sensi. So, and also that's supported by a pretty, pretty elaborate, um, LAwareN network of sensors to add a little bit of context when we're lacking. Some of the data we need to, to onboard our assets. And, you know, like our steel mill have, um, the complete array of age of equipment and complexity of equipment from the very, very simple to the, the latest and the most modern equipment. So we, we're dealing with a vast variety of equipment and, you know, everything from, um, you know, Bringing in iron ore and raw materials to completely finished products going out the gate to our customers. So it's pretty, uh, uh, when I say steel mill, it's the whole thing from front to back. So, yeah, it's very cool to be part of that. And, um, you know, uh, I spent a long time just in one department and now I sort of pop up every around the place and try to understand what the needs are of, of that particular business unit and see where Sens I can, Most quickly help them out with what they are, what their sort of pain points are, so that's that's a bit about me and what I do. Fantastic. No, really good, um, really good good background there, um, Chris. um, and I guess, uh, you mentioned, you know, the scale, you know, the scaling of the platform across all those different sites, um, globally, which is really fantastic to hear. What, what would you say has sort of been the, been the key to that, you know, to that global rollout? It doesn't always happen that way. I guess the, the success stories always spark people's interest. And that they kind of multiply when, when people sort of, sort of hear some of the success stories we've had and they're like, Oh yeah, I have similar equipment, I have similar needs in my department, or it's completely the other way, where people are like, Oh, that's cool, but can you do something with this? And um You know, It, it's, it's still all about people and engaging them in the right way and, and finding the, you know, every, every department will have those 2 or 3 people that sort of naturally jump at the chance to do something like this. And it's really, it's finding those people and, and getting started with them. And that first sort of couple of success stories that come from, you know, their own department, and that's when the more people like from that particular department, they jump on, on the, the hype train, I guess you'd say, and, um, It just multiplies from there and then, you know, we, we start to, to talk about our latest sort of thing is, you know, to increase the rate of onboarding, we're, we're trying to find as many people as we can in the department to onboard, say, 5 assets a month. So if you start to, you know, if you find 345 people, and then, you know, You expand their capabilities to get them up to 5 a month, and then you add more people to that group, and then you add more to their capabilities, they can onboard more assets, you know, per month. So, that's the latest theory on how we're trying to, to roll out, um. So it's a big, it's a big push about training and upskilling our people and in giving access to the platforms, training them up, and uh yeah, the, the LoanAN network is, is handy, particularly on some of the old, Uh, the older, less critical sort of equipment, like we do our critical things quite well, um, we find it's sort of that next layer of criticality, not the ultra like important stuff, it's the next layer down that can not, that doesn't get as much love as the critical stuff where we can really start to make an impact, uh, on, you know, detecting anomalies and, and trending. Sort of metrics that we're interested in and yeah, that's where we can make most of the impact. And that's, you know, like, and that sort of maybe frees you up to do the, the critical stuff better, and, and not to say that we can't do anything with the critical stuff, it's just, you know, it's critical for, for a reason because it's probably a little bit more complex, um, and it, it takes a little bit longer to, to understand how we can implement Sensi sort of in, in those more complex sort of processes or, or machines. So that's, that's generally how we try to do things. Yeah, and is it generally, I, I guess the general point is it's about building momentum internally. I mean, or, or getting on the, and people don't want to miss out if something really positive's happening internally. So how important is sort of sharing the successes internally and widely? And how do you do that, I guess, as well? We have so many different like, um, Steering committees and focus groups, and, you know, we have groups of asset managers that, that meet regularly and there, there's so many different ways. We have local intranets we can push out, but it really is those, you know, those one pages. I think you'd probably come across, we send a lot of them through to, you know, our, our day to day sort of Sensor contacts and Semen contacts that we, we report into, and, and I'm not sure where they end up in, in your organization, but that's a really good way of sparking interest. And then, um, You know, and, and those people that, it's always easier to go to people that are, are interested and, and they have, you know, a few, a few people that have got some spare time and they've got a little bit of spare capacity and money to, to kick it off. Um, it's really hard to ignore those people and they might not be the, the areas that we've been asked to focus on. So there's, there's, there's critical parts of our business that we need to, you know, improve reliability or capacity because of, you know, expansions or, or whatever we need to do. And, But then it's the, so we've got to go to the critical people, and then there's sort of the um The next group of people that sort of haven't been deemed as our critical departments that need the help, it's the other ones that have, you know, maybe a will and, and a few spare people that we, you know, you can't ignore them because they're, they're so enthusiastic and they're so on board, and they want to jump on the hype train and, and get things going in their department. So it's trying to, it's a bit of a balancing act, but, you know, we're, um, our, our small team, and it, it is a small team, it's growing at the moment, so. That's good, and, and, and those sort of things maybe toughen up in the, in the steel market, it's, it's one of our key focuses is to, you know, roll out predictive maintenance and save money through routine maintenance and also avoiding those, you know, critical breakdowns that, you know, cost you time, money, And lost production, all those types of things, so it's, uh, it's definitely central to our business now. Yeah, so that, that is interesting you mention about, you know, what do we, you know, we as Siemens do internally. So we do, we do share them internally because obviously we want to celebrate successes with customers. And I think the people are amazed by some of the, especially with, you know, the Bluescope Steel case, case, uh, or projects. Um, as an example, because some of the figures on there, almost to, to some people it was like, wow, is that, is that really true? kind of thing that is, it's a common thing with Sensa generally, is like, is that really true? Can it really do what it says? But do you encounter that kind of sort of skepticism internally when, you know, when you see these great successes and, and numbers and what's going, going on in the project? Well. We're, we're very cautious about, you know, Overestimating, and, and, and it is a lot of estimating in that, when you, when you sort of avoid some sort of, um, breakdown like the one we're sort of talking about today, it, it is a little bit subjective, but we always are quite conservative when we, when we do estimations, and it, it's also based on past experience too, like, you know, this particular one we're talking about, this hydraulic success case where we avoided 24 hours downtime on a metal coating line. Um, we know that that is, I mean, it's still conservative, uh, because of the, the potential ramifications of this particular failure or potential failure. Um, So we are always very conservative and, and we report this into the highest level of, of our company, so we, we always, Um, have in the back of our mind that we need to back up our estimations with, with some sort of, um, facts and figures, and, you know, generally we don't do that ourselves as a predictive maintenance group. We rely on the, the local departments to, to sort of make those estimations, and we're always very, very conserved in that space. So, um, You know, we're sort of always open to scrutiny. And the second that, you know, that credibility is gone, then it, it's gone forever. So we're always very, very conserved in that space when we make estimations and, um, it all, it's, it's based around, um, sort of lost potential lost revenue is most of the things that we calculate, but, um, now sort of our, our general management and executive are pushing us towards, um, well, well, now that we've got this, Sort of monitoring in place on certain assets, the next step is to start to roll back some of the maintenance, the routine maintenance we have. Uh, and, you know, you need to have a little bit of, you know, longevity in the app and, um, confidence in, in what you're seeing that the app's telling you to, to start to do that. That's when, you know, you start to get brave and turn off those maintenance plans that have got us to the point we are now, and, you know, it keeps, Keeps things ticking over, it's, it's, yeah, that's, that's when you really know you're cooking that, you, you know, you've got the confidence of the local maintenance practitioners to start to turn off those maintenance plans and, and rely on what the app is, is actually telling you, so, That's, that's a big step. It's definitely a big step. Yeah, it's that, it's that element of trusting, getting to the point of where, you know, you trust the app and what it's, what it's telling you, right? And that's, and that's obviously a very key point of for adoption or widespread adoption is. Uh, it actually is, and, you know, it's, it's those, you know, you might save 2 or 3 maintenance people from coming in, uh, Saturday night for a Saturday night breakdown, and no one likes doing that, right? Like, no, everyone hates that. And that's, that's one of the things I really want to save people from doing those Saturday night discos where, you know, you get a phone call in the middle of the night, come to work and fix this. So we're trying to, you know, make sure that doesn't happen, you know, and when you save people from having to do that a couple of times, they, they, they're on board and You know, but it, it takes a whole bunch more than that to, to get a whole department on board where to the point where they're like, OK, we're going to start rolling back some of our maintenance plans and, you know, fully trusting on, trusting, and we're, we're all in on this sort of process. So that's the, um, that's the battle that we're, that we're, well, I think we're winning it, in, in at least most of the places we're, we're rolling out in. So, yeah, it's, it's very encouraging. Very encouraging. Yeah. And just before we we go into that specific case, um, success case we're gonna talk about today, um, maybe Chris, you, you explain a little bit more about the, the KPIs. So you mentioned a few, you know, the KPIs that you'd maybe report upwards or that, or the, or that you and the global teams track. What, maybe you could go into a bit more detail about that. So when we first started, first started, uh, rolling out and within the, um, the trial period, we would reports of downtime avoided, you know, like you'd, you'd, you'd see a, a problem with a, a, a specific machine and you'd, you'd manage that problem until some, um, Predetermined downtime, or you, you manage it to the next shutdown or, or, or whatever, like, and that would be, OK, we saved, you know, 12 hours downtime by managing that blower or that motor to that, um, maintenance window. We take care of it then and there, and, you know, we've avoided a, a line stop, um, an unplanned line stop and some, some downtime. So that's where we sort of started out. But maybe as your, um, production units become less loaded and there's, there's idle time, then, Downtime avoided kind of becomes useless, so you sort of change your thinking a little bit into the point where, OK, So we have, you know, maybe detected a problem with a machine, a pump, or a blower or a motor, and You know, we've We've pulled it out of service and replaced it with a refurbished or a new unit, and You know, that repair on that motor because the bearing, you know, we detected it was hot and we managed it to the point where the shutdown was, but it kept running and the bearing didn't collapse, which means we didn't damage the shaft or the housing, so the repair is, Is a lot more efficient than or cost efficient than than what it would have been if we wouldn't have detected that, and we had a an in-service failure. So we're, we're trying to, instead of the, the downtime avoided, we're like, OK, now we're saving money by, you know, a repairing equipment in a much less sort of failed state than, than what it would have been if we didn't have the monitoring in place. So we're trying to measure that and we're trying to measure like well what's the, um, What's the, the gain of offsetting maintenance plans, like either changing the frequency or, or shutting them down altogether and just making them like a, a manual call when, when the app tells us that, OK, this, this asset's displaying some elevated temperatures or vibrations or current metrics that, that we haven't seen before, and now is the time to sort of, OK, let's, let's do a, an inspection at the next, you know, maintenance window. So, we're trying to calculate, Offset savings from reducing or shutting down maintenance plans and, You know, there's sort of a general rule of thumb if you do something in a planned fashion, you're going to get the most efficient. You know, labor cost outcome. In that space rather than doing it, you know, after hours or in an unplanned type fashion. So there's, there's savings to be, to be had with doing things in a planned fashion. Um, in theory, the, the less repair bills and, you know, also shutting down of the routine maintenance, and we only execute instead of time-based, we're doing it, um, when basically Sensai tells us to, so. You know, and if you sort of gear that towards, if you start at the, the things that are frequent and we, and we're using outside, you know, labor, if you sort of start there, they're the, they're the sort of big money winners if you start in that space. But also there's, there's thoughts of, you know, internal labor is still valuable and, and we always have more work on them what we have labor available, um, So there's still money to be to be saved, and I guess how we think about it is, you know, we're We're sort of offloading some of those routine inspections or line walks or whatever and we're making our people, We're freeing people up from that to, to redeploy them to more critical, high-level things than just doing some, you know, nice, nice to do, nice to feel, um, inspections that, you know, 9 times out of 10 don't really tell you anything. So that's the latest sort of metrics that we're looking with. Yeah, I think, I think that's really interesting actually how you say it, you know, like you said, it's, it's evolved from looking at things like downtime because we often say with, with Senite, it's actually, you know, looking beyond, we encourage people to look beyond downtime as like the metric to, um, to measure, um, success. So that, yeah, so, so that's really, um, really interesting. But, um, if we, if we look now at the, you know, the specific case we, you mentioned before, so the hydraulic system. And maybe I guess first of all, Chris, we could just go into a bit of a brief overview of the, you know, the background to it, you know, where, where it's located and that, and that kind of thing and what was happening with it. But some of the, some things get left behind because that's, you know, you, there's a certain amount of engineering time and money and resources to, to upgrade and you wanna, you know, obviously you look at bang for buck type stuff and, um, anyway, you get into this space where, you know, a few bits and pieces get left behind and, and maybe, This, this hydraulic system that used to just, you know, hold a few snubbers in the air and when the line stopped, it'd lower a few snubbers and, That would be, you know, it'd be sort of a low criticality, doesn't really matter if it stops type, um. Piece of equipment And as you sort of upgrade and modernize things, I think there was a big movement to go from like water-based hydraulics. Which would service most of our exit ends around our, our recoilers and coil cars and stuff like that. Anything with a really big cylinder on it, we would use, um, sort of water-based hydraulics on and anything a little bit smaller, it would, it'd be sort of mineral-based sort of equipment. …

Full transcript (WebVTT)