Michael Desabris with IBM

SUMMARY KEYWORDS

maintenance, reliability, asset, apply, companies, palo alto networks, maximo, technology, fmea, solution, ibm, industrial, ot, linkedin, c suite, years, people, perspective, condition, library

00:00

Industrial Talk is brought to you by Palo Alto Networks. You've heard me talk about my friends at Palo Alto Networks zero trust OT security delivering a comprehensive security solution for all OT assets, networks and remote operations. But did you know that the Palo Alto Networks solution provides over 1100 app IDs for OT protocols, over 500 profiles for critical OT assets, and over 650 OT specific threat signatures, now that's best in class security delivered, learn more about Palo Alto Networks, zero trust IoT security solutions, and how you can achieve 351% ROI over five years. Learn more about the Palo Alto Networks solution by going out to Palo Alto Networks.com. That's Palo Alto Networks.com.

00:56

Welcome to the industrial talk podcast with Scott Mackenzie. Scott is a passionate industry professional dedicated to transferring cutting edge industry focused innovations and trends while highlighting the men and women who keep the world moving. So put on your hard hat, grab your work boots, and let's go around

01:14

once again, welcome to industrial talk. I truly appreciate your support of a platform that celebrates you industrial professionals all around the world. You are bold, you are brave, you dare greatly you innovate. You are changing lives and therefore changing the world as we speak. Thank you very much for what you do to make the world a better place. We are also broadcasting from SMRP 31. That's 31 years of incredible solutions being presented here. It's fantastic. And if you're in the world of maintenance, asset management, and reliability, whatever it might be SMRP.org is the place you need to start go to estimate SMRP.org so that you can get that ball rolling in hotseat, we have a gentleman by the name of Mike and he is with IBM. How do you say your last name it DESABRIS?

02:08

Desert breeze? Say that again? Desabris?

02:12

Desabris. The S is silent. Yes. Silly me. French is, of course, silent. My wife speaks French. Okay, of course I don't.

02:25

I don't either. So.

02:28

But you have less than

02:30

you haven't a good conference on great conference. Yeah.

02:32

Yeah. What makes it great.

02:35

I think this year is record attendance. Yeah. Right. So glad to see everybody back out again, glad to see the enthusiasm around reliability and maintenance. You know, when new people come in with the challenges and what they're looking at, you know, I work for IBM Maximo. So very, very excited about the product innovation that we're bringing to market, how that's fit in, fitting in with challenging problems that customers are trying to solve. I love it. And all that coming together. And it's been in the works for probably about six or seven years to bring basically what I said, putting a brain on Maximo. And now reliability into the suite. So it's a complete suite that companies all over the world can see,

03:17

this is what's interesting. Yeah, Maximo, it. It has so much functionality and a lot of a lot of companies that can enhance it and bring out some greater value in using the solution. That's amazing. What what is, what I get a sense is like, I've been in industry for many, many years, and, and I've been a part of organizations that talk about, hey, we want to be a Reliability centered organization. So we're gonna really focus in on our assets. And then one year later, the the devices are still sitting in the cubicle and they're still reactive. And and I get a sense that we've turned a corner from from that perspective, that asset management, it seems to me that the C suite is finally getting it. And that these are truly not sustainable from a environmental point of view, but sustainable from a strategy perspective, initiatives that I think that that's important, right? And I think that the

04:24

I see that too, so I think really want to change probably over the last five to seven years. what technology has done with the C suite, as we brought machine learning and algorithms and analytics, where now the C suite is looking at it and made me unsexy. Right before maintenance was always the redheaded stepchild in the back room. So now it's on the forefront. We're gonna do predict technologies, right. Yeah. But once they go into, like you said, it's not a tool. It's not something you just buy and you install and obviously you go from reactive to proactive. It's a Culture change, right? You're gonna have that type of culture, reliability culture, your quality culture, you know, similar as like a safety culture. You know, it takes time for the people to change and the organization change with the leadership, but because now we have this impetus of technology that's getting simpler and simpler at a very, very rapid pace. Companies have seen the benefits in the business case of actually doing that transformation and want to engage their organizations in that transformation. So that's where I see a lot of traction is that, you know, maintenance reliability, because of those predicted technologies. And now steps, right? In the business cases, yeah. So you can prove out the business case. And if you can increase your uptime, by a certain percent, reduce your costs by certain percent, and you put that across 10 plants.

05:48

It's like, there's, from an operations perspective, there's always low hanging fruit, you're always gonna get in the beginning, you're gonna get some really, you know, wow, results. And that's really great. Got it. But then after you remove all of those inefficiencies, then it becomes more mundane, for lack of a better term. And that it's not as if you're not operating efficiently. It's just that you've removed all of the inefficiencies. Yeah. And then that's where it starts, from my perspective, the C suite says, Oh, why are we doing that? We're super efficient. Right? And and we're not achieving those massive,

06:29

you know, gains. So efficiency and effectiveness are two different words, right? So the effectiveness of your maintenance program is you're reducing, or eliminating downtime, right? If you're fixing the same problems over and over again, you can become really efficient at work or the management can be able to fix it faster. But that's not the point. The point is to eliminate that failure at its source. So you don't have it in the first place. So really, maintenance is about not doing maintenance. That's the goal. Yes, yes. Right. That's the goal, you don't want to do maintenance. That's the whole idea of it. And that's where now the technologies are so advanced. And they're now getting so easy to apply, that you can actually take good leaps, leapfrog steps and big step function change across your organization, as you're changing the culture, right.

07:23

Now, I'm going to walk away with the statement, you don't want to do maintenance. That's the goal, you don't go. Because I had this conversation with a number of other companies. I said, yeah, it maintenance, the spin maintenance spend from a financial perspective is a one for one. So yeah, whatever, you know, at the top, you, you know what that $1 that you had to use and maintenance is $1 off of the bottom line, because you don't get to depreciate the, the expenditure. And so that's always been the drivers sweat the asset as much as you possibly can, for Goldman maintenance as much as you possibly can, and then do maintenance at the right time, like at the optimal time now, where you're getting a catastrophic failure, but you just like, right, they're gonna peak,

08:15

whatever that is. That's where that's where the technologies around monitoring and around predictive or anomaly detection on condition based maintenance, I'm mature enough now that you can apply them efficiently, effectively and cost effective manner. So that you can go through that condition based approach, and then focus on when is the optimal time to do that means, right, I know I need to do this. But now if you choose SMRP curves, if I go up that PDF curve, how far can I get up that PDF curve, and then traditional condition based rate and it's only going to say up to a certain point on the knee of that curve. Now we can go much further with these advanced analytics, because we're looking at that time series data, IoT, IoT over time, and then we can actually start to predict or actually know when something has changed. So you're managing more by exception, and then guessing when the right time to break in and do maintenance or the line went down, let's do this extra maintenance that we have and things like that, then management's also looking at it and becoming more of a partnership with the maintenance related to building the organizations because they realize if I can increase my uptime and my throughput, the money that I'm investing in the asset is not a cost. It's an investment. Right? And if I'm proactive about that, and if I do it in a way that's based on reliability strategies like product, it can be very very effective and efficient at the same time.

09:44

Yeah, I would always fascinates me as is in a short period of time and I mean, really from a short period of time, the the the innovation the technology has has rapidly caressed, I don't know, I still believe that we're at the tip of the iceberg. But this is really quite a renaissance from my perspective of, of the attention that it's getting, right? Where do you see this going? What's like that mean? Put your future hat on.

10:22

So the future hat. Let's talk a little bit about what IBM Maximo we're doing now. That's bringing the future today. So we just launched a product called reliability strategies. So it's just grinded just strategies, strategies. And so what it is, it's a combination of an RCM tool. And it's also the FMEA library that comes with it. So it's about 800 assets. So in the past, you would do these studies, it would take you four weeks, five, six people, your best people cost you a lot of money, these people don't have a lot of time, because they're important. So those studies never really get done. Now in an accelerated like this library, you're applying the FMEA, you're not creating the FMEA. So you can apply an RCM study 75% faster, and get those same all the time it's report benefits. Now, what that allows you to do is go further into your asset base. So you're more proactive. And because it's based on a failure mode, you're doing the right maintenance at the right time, and then applying the technology to support that.

11:29

Is your speaking at the Innovation Center. Is that what you're talking

11:33

about? That's part of it, and that at the Innovation Center, what we're doing now, you've heard a chat GPT. Who,

11:39

who hasn't? I mean, what have you been living under a rock?

11:44

She's so if you look at if you take that, let's apply it to reliability, right? Where does it fit in? No, it's not there, that's going to take someone's job away. Right? It's not going to do that can make really everybody more efficient, effective. And what IBM about 30 of our companies still aren't getting is about 330,000 people. So we're in our old school where we had a certain budget for r&d set aside, we still do that we have 100 year old company. And that's how we come up with these innovations. So we're applying those foundational models and check TPT models to build out this library. Right, this FMEA labor, so we're quickly going to go from 800 to 8000 assets. So as Mr. Customer, you're going to be able to apply these quickly and have all your asset bases covered. Right. So now you're going to be doing a couple hours, what would take you those six best people three or four weeks to do you have power that the downstream effect is that as your PMS get optimized, and then we start applying advanced algorithms. So if we know the failure mode of the asset will say its vibration. And we say okay, one of the mitigation factors is we're applying the sensors in this algorithm. While currently, a lot of those algorithms have to be developed by data scientist. So we're also taking that same type of foundational model, and rapidly building out the algorithm to match the failure. So in speak, if you can always look at when technology meets, you know, the domain experts. So the reliability just saying, Well, this is the failure mode. And then the data science will say this is the parameter. But they're actually the same thing. So now we can connect those look at IoT IoT type of streaming data, and automatically apply that model that was built by the foundational model, not by human. Right, so now we're cutting to reliability, I mean, the data science side of the equation, so the whole full circle, three years ago, it would take you that amount of time to do the FMEA study, I've done this in large auto manufacturing companies. And it would take you two months to set up and do those type of knowledge interactions, right? Now, you're gonna be able to do it in a matter of couple of hours. No kidding. The scale of that, right. So, so what would always happen when we would do these type of pilots for companies, they'd be like, okay, great, you're proved that works. And you prove that one station, how do I do 900 simultaneously? So that's where we're gonna see the scale, the rapid adoption, rapid implementation of tying the basis of Reliability centered maintenance with FMEA is too advanced.

14:30

You FM be a very

14:33

emotive affected.

14:35

There you go. Thank you feel your blood and in fact, that's okay. I might be speaking fast. No, no, no, no, that's okay. Because I, I'm very sensitive to acronyms. Make sure that the listeners understand. Very cool. Yeah. Now that that exists today. What do you see going? Well,

14:51

once that starts to the once those foundational models get embedded in the solutions, that scale is just going to become More and more automation. Yeah, right. So there's gonna be less and less effort by companies to apply it. So the C suite that kind of looked at, hey, I got this shiny object called predict a few years ago, I just plug this in and all my maintenance problems go away, we're really only addressing the part of the problem. But now you're going to be able to apply that and mortar assets quickly across the board, and be able to do that in a way that is cost effective.

15:28

With that said, how do you take in consideration the variability, let's say, motor a, this is the same motor a, that is down south versus the motor a that is in May, right? The conditions are, you know, different. How do you take that into consideration? Question?

15:53

So that's built in as part of it, called the operating context. So we look at the criticality asset in combination with the duty cycle, how often you run again, right? If it's a parallel pump, once a backup, you're only running 50%, if you're running a full 100%. And then you're looking at the environment, which he writes his own harsh condition, right. So if you see your car, if you just took it to church on Sunday, versus I'm going to take it out into Baja, and run the hell out. Right. So it's that way we look at that operating context, and you can select those as you go through applying the study, not creating the study, applying it for your operating context. And the changes.

16:36

Is the is this information accessible, consistent among for anybody? Let's just say, company A, company B, I'm accessing this information, this this, this equipment list, whatever, it's, it's, it's available? Yes. Because it would be impossible to create multiple for different companies. It's got to be just a common sort of a location, right? It's a library. It's a library.

17:08

It's a cloud service library that supports Yeah. Wow, a library, a library that's reusable and scalable. It's where everybody wanted to go, but didn't have the solution. Or the investment to start.

17:23

See. I really I think that that's just I don't know, man. It's it's exciting. Is it it is going to continue to grow. Oh, yeah. So it's not static. It's dynamic.

17:34

Its dynamic. As more and more people get involved.

17:38

Yeah, just hones in on it's like my gosh, it's just fantastic. To get a hold of you might they

17:45

can contact me on LinkedIn, Michael Deseret. Get me sa VR is responding. It

17:53

is. I have his name. I have his contact information to you act amount on LinkedIn. You active out on LinkedIn.

18:02

I go in spurts depends on my schedule. Well, we're gonna have his

18:07

LinkedIn contact out there just because we want to know a stat card what it looks like, even though it might be a little dated. All right, thank you. You were absolutely wonderful. Thank you. All right, listeners. We're gonna wrap it up on the other side, we're gonna have all the contact information for Mike out on industrial talk.com. Remember, go to SMRP.org get engaged, be a part of the organization, you cannot lose on that decision. So go out to SMRP.org. They've got symposiums, they've got other events. It's an exciting time to be in reliability, asset management and maintenance big time. So stay tuned, we will be right back.

18:46

You're listening to the industrial talk Podcast Network.

18:55

Mike Desabris, IBM is the company. That's SMRP was the event. Now his LinkedIn stack card goes by Michael so just make a note of it. But don't worry. So I don't industrial talk his contact information. Everything you need to know about Mike is out there. Connect with them. A lot of great work being done in the world of Maximo, IBM, all of this stuff to help. Reliability asset managers, maintenance professionals do a better job. Excellent, excellent conversation, put SMRP on your calendar. That is a must. If you're in that profession, that is the place for you to be go out to SMRP.org and find out more they've got a ton of stuff for you to get engaged. Industrial talk is here for you. We have we have education we have podcasts we get amplify your voice you reach out to me go to contacts just right there. Let's chat and you'll be talking to me. All right, be bold, be brave, dare greatly hang out with Mike and you will change the world. We're gonna have another great conversation. Shortly