SUMMARY KEYWORDS
data, industrial, analyze, models, world, reliability engineer, tools, good, goldstone, palo alto networks, reliability, decisions, john, find, give, ai, assets, trends, set, machine learning
00:00
Industrial Talk is brought to you by Palo Alto Networks now Industrial Talk has a direct link to a new report about Palo Alto Networks industrial IoT security. The new report analyzes improved ROI and reduced complexity. The Palo Alto Networks industrial IoT Security provides customers significant savings and benefits in the following way. faster time to industrial IoT protection, reduce operational complexity, and reduced risk to the organization. Find out more how Palo Alto Networks industrial OT security helps you better manage, secure and protect your ot assets while providing safe access to the assets and services required to modernize your operations. Go out to Palo Alto networks.com. download your copy today, and begin lowering your risk and achieving your 351% ROI.
01:04
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.
01:21
And let's get like once again, thank you very much for joining Industrial Talk. And thank you for your continued support of a platform that is dedicated to industrial professionals all around the world. You are bold, you are brave, you dare greatly. you innovate, you collaborate, you solve problems, and you're making the world a better place. We thank you very much for what you do. And that's why this platform is dedicated to industry professionals all around the world. We are broadcasting from the 31st annual SMRP conference, right here in Orlando, Florida. It is a collection of problem solvers. That's what it is and people with solutions to help maintain help asset manage your business even better. They're all here. If you have an interest in your in maintenance, assets, management or reliability, go out to Ian SMRP.org. Find out more be a part of the voice. And you get to connect with individuals like John Todd, how are you doing, John? I'm
02:13
doing very well. The end of a long day.
02:17
Oh, I hear you. You having a good conference? Oh, yeah. Yeah. Let me tell you how long you've been part
02:22
of SMRP. This is probably the fourth or fifth year that I've been over the years. Yeah. So not not 20 years, but not one either. So yeah. Yeah, that's Kevin.
02:33
And yeah, he's he's been around for a little while. That's for doggone Sure. And he knows a lot of stuff just like you. So with that said, give us a little background on who you are. Sure,
02:42
sure. So I have a degree in electronics engineering from days gone past, you know, hardware degree. And I've always worked for software companies. So ever figured that one out? I've worked with as a reliability engineer back at the Deep Space Network. A few years ago,
03:00
look, Hold on What do you mean what do you mean by that? The
03:01
Deep Space Network? Yeah, the the the antennas that track all the spacecraft. And such. That was a lot of fun. Where Where was that located? That was in Monrovia, California. Worked with JPL and such
03:13
did you go out to Barstow? Did you go to Goldstone? Oh, yeah, all the time. Oh, there you go. I
03:17
love it. Yeah. Yeah. Did you know yeah, we take tours of the goal. Yeah, yeah. Yeah. Okay. Like you, I would get to go out there all by myself. So I didn't need to know where? Oh, yeah. Oh, yeah. That was a lot. That was a that was a great gig, because we're just working with cost. Absolutely one of a kind equipment, right. And, but the precision. And the reliability that was built into it was was pretty amazing. And of course, as reliability engineer, my job was to tell all the mission managers and such as like, the antennas are going to be good for your mission. No problem. That was a lot of fun.
03:56
That's exciting stuff. That
03:57
was that was I don't
03:58
know how you're gonna top that conversation. I could just grip on that for a little while.
04:04
Yeah. All right. The data we had was just really hard to work with because everything was custom. Everything had their own format. And we had to bring all that and filter it together. And now I look at all these machine learning tools. And
04:20
I wish I had that. But you're probably doing it now. Or they're trying to imagine they're, you know, they have to because it's it's just a tsunami of data. Right,
04:28
right. Exactly. Exactly. Yep.
04:30
All right. Continue. Do you have more detail? Yeah, just
04:33
a little bit, just some quality Quality Management Process Management for an Air Force contract out in Colorado for quite a while and done a lot of risk management type thing and all along the way. I've dabbled with Maximo. Yeah, of course. Right. Yeah. And so now I'm working for total resource for the last just over eight years now. So okay, and of that liaison between the salespeople in the technic Oh people and so having a good time with that
05:02
is a is a is TRM. A Maximo
05:09
vendor or that's our primary primary focus is business partner for IBM. Yeah, for Maximo. There you go. I knew
05:16
I was like, You got a relationship? Yeah, I didn't know how to put it together around. But clearly I failed miserably. Yeah. So let's talk a little bit about the topic you want to talk about. That is that's, which is really interesting, because what we have is, is this continued focus or increase or hyper focus on data is just, it's like, where the gold is located. And everybody's talking about it. And everybody wants to analyze it, and everybody wants to analyze it in a way that that is meaningful and tactical, or whatever it might be. Help us understand that a little bit.
05:51
Sure. So I had a friend of mine, who was a statistics professor at a college years ago. And he said, When you are presented with data, don't do anything with it. Just look at it. What does it tell you? And you know, at first you look at a page full of numbers, and you can't tell anything, right? But if you really concentrate it, so you know, there's a lot of threes in this dataset. Oh, I don't see, there's only like one number seven. Right. And so it's, that will just kind of give you a sense of of where you're headed. Right? What you would kind of expect. Now, take that idea and look at gigabytes worth of data, right, coming in every fraction of a second. You just can't humanly do it. Yeah. So we we then apply tools, we try to use tools, you know, some are machine learning. Some are just basic, you know, models, and they tell us things. But does it kind of match our gut feeling? Right? Do we just say, oh, yeah, that's, that's saying this is going to happen? And I believe it? Well, we need to bring some decision making into this. Right, we need to Sure, trust the results to some extent, but to question them as well, you know, to look back and say, Can I, you know, can? Does that seem right to me? Right? So there's so much data now, you're absolutely right. Just to go back to your first point. It's this wave, right. You know, we used to have data warehouses. Now we have data lakes. Yeah. Now we have data lake houses. Right? There's another term, right?
07:40
No, I was either gonna go on, is there another one? I'll put their link there is and
07:43
there'll be another one in six months, right? You know, we said not too long ago. You know, storage is cheap. Data storage is cheap now. And that's still true. But we're storing terabytes and terabytes, petabytes worth of data. Now. So but the basics are still true. You know, there's patterns in that data. There's information insight, right?
08:06
Yeah. But see, this is an interesting, this is a challenge, because you'll see the patterns. It's like, you're looking at that 3d picture. And you gotta blur your vision a little bit to see things just pop out. Right. But where is where's it? It never cuts off. You're always analyzing the data. You're, you're you're constantly being pressured into finding that that that additional gold Yeah, it never ends. Yeah,
08:35
yeah. And the moment you analyze a set of data and say, Here's my results, here's my, my confidence in whatever it's telling me. Well, that's so five minutes ago, right? Because while while you've been doing that, there's another flood of data coming in that you'll analyze next week, next week, next week. Right? So it, you're right, it is never ending, and it's going to grow. So the the power of the tools that we need to use to make sense of all of this needs to increase as well. Right.
09:12
So yeah, absolutely. So take us through something like that. What do you what do we what do we use in to help help that that happen?
09:19
Sure. So every, you know, the, you can't have a conversation anymore these days or do a presentation unless you use the term AI? Right?
09:28
It's, it's becoming the miscellaneous file like industry for Dotto, right, exactly. All of a sudden, everything is like yeah, today, I heard
09:36
some people talking about industry 5.0 the other day and I said, Wait, we never finished three.
09:41
I know I heard the same thing. And it's like, okay,
09:45
but let's talk about machine learning. Right? That's where we really get some some utility out of all this. So these these models that are being built Right, to ingest whatever the source is a data, you know, and multiple channels of data, right, not just, you know, a couple of variables, but 10 variables, 15 variables, going into these models and the model understanding, and like you've heard previously, you know, what is normal? Right? Or what is not normal? Or are we trending towards not normal, those models being able to ingest all that data, and then start to give us those insights. Okay, that that's the the real power right now. You know, not necessarily, you know, yet making decisions for us right now. Now we're in the AI side where it's taking actions based on probability. Okay. But, but these the the outcome, I want to say, the, the results or the outputs of these models giving us insights that we perhaps never had before, because they are able to look at all these different channels of data coming in and find patterns and trends and, and things that we just we just could never see. Right?
11:10
Are we in a position to make better decisions?
11:14
I think we are. I think we are, as long as we as long as we understand the model, right? What, what it's filtering out because they all filter things out, right? We always say, Well filter out the noise filter out the noise, get right down into the good data, right? Well, you know, what, sometimes you throw away sometimes there's good stuff in the noise. Yeah, right. Yeah. So we shouldn't be afraid of the noise anymore. Okay, let's accept the noise, because maybe there's trends and patterns and insights in that noise that we, you know, just ignored in the past. So given that, yes, we could make better decisions. We certainly could, because we we should know more.
11:55
Right? But again, it's where do we get to a point where, where something is completely an ultimately optimized, we just have super-duper vision and clarity into the operation. And then when you reach that pinnacle, whatever that looks like,
12:12
it'll never be there. It'll never be ever be there. And I'll give you a good example. So the sprinkler system at my house, the controller has an AI feature in it. And it just says, oh, we'll take care of you know, I don't have to set my schedule, you know, every three days and 15 minutes in every zone. I don't have to do that. I just say, let the AI take care of it. Okay. It does a terrible job. Terrible job. You know, Oh, you don't need water for 10 days. And so in the middle of July. Yeah, right. Okay, wrong. Don't think so. Under the covers, for each for each zone. In my house. There's 18 different parameters that I can set the soil, you know, how much how much sand is in the soil? How much shade and on and on not? And I go, Oh, I'm tweaking the model. Right. Yeah. So out of the box seems like a good idea to say I take care of it. But I need to have an understanding of each of those parameters so that I can set them better properly to match my little world in my house. Yeah, right. Now, the machine will learn that, you know, it knows the local weather, right? So it brings those factors in as well. But ultimately, it's making a decision to turn on that sprinkler zone or not. Right. But I'm intervening. I'm setting that up. Right. So who's making the decisions? Right, yeah, yes. Just look at the data. And so they'll yell everything lines up, turn on the sprinklers, right. But I decided how those parameters are set to Mantis thing. And my contact did it did doing a better job. Yeah, it's still not. I still like my regular every three to every other day kind of thing. It seems to work a little better, right?
14:04
That's how I have mine set up is like, I have a rain gauge. So if it rains, right, it doesn't. It's like okay, so I
14:11
can yeah, there's one channel. Yeah, that's right. Yep. Yep. Thumbs up.
14:14
I'm all good. Yep. But if it doesn't, right, I want that thing on. That's right. That's, you know, so what do you think it goes? Where do you think you're? I mean, we're still at the tip of the iceberg. Oh, hell yeah. Yeah, definitely. Definitely. We're just, I mean, all of a sudden, it if it wasn't for Chet GPT. And all of a sudden, that becomes the sort of the latest lexicon, right. Yeah. Right. And applying it. Where do you see it all going? So
14:39
ultimately, it's really just another tool, right? I mean, everybody got all excited when spreadsheets showed up in the world, who I can do calculations Right, right. And then and then the idea of databases came in to play right relational databases, oh, I can store all this data and then all the analysis till you know the business A business intelligence tools and things like that charts and graphs and, and all of that, and really the machine learning approach. And then ultimately I, you know, it's just another business tool, right? It's very powerful business tool. But if we can learn how to apply it, right, and understand it better in our context, whatever our business is, you know, manufacturing or forestry or facilities, management, whatever, and then understand these the construction of these models, then we can make some decisions based on those results, right? It's just, it's just another tool, honestly. Yeah,
15:41
but everybody's looking for that silver bullet. Oh, of course, you know, everybody wants this, like, I have a problem. I need something that I just flip a switch and boop. Right. Alright, good. And then, you
15:51
know, in some cases that may come right now go back to the sprinkler says, Yeah, right. It could probably do a really good job, once a little bit of justice, I'll just let the thing run. I don't have to worry about it anymore. Right. So there's the application of these new tools. I thank you, right, we've we've kind of just scratched the surface of what they can do for us, because I
16:15
still think that there's room for blocking and tackling the standards. The when you start looking at reliability and maintenance issues, there's we're talking tools that say, Hey, there might be a problem there. Or there might be some challenge, whatever, it might be a notification, but we still need people to go out there. Yeah, analyze it. And yep,
16:34
yep. You know, there's that prediction aspect of that we're looking forward to you know, tell me what's going to happen in 20 days. Right. And they're getting pretty close. So I like that. Yeah.
16:45
How do they get a hold of you, John? So
16:48
work for total resource management. We have a YouTube channel, pretty easy to find. And John dot Todd at TRM. net.com. Very
16:56
good, my friend. Thank you very much. You were absolutely wonderful. Thanks, Scott. All right, we're gonna wrap it up on the other side, we're going to be having all the contact information for John out on Industrial Talk. So if you're not you'll be able to connect with him. You active on LinkedIn? Yes. Okay. Yep, we'll have his LinkedIn stack card as well out there on Industrial Talk.com. We're broadcasting from the 31st annual that's MRP conference, put this on your calendar for next year, if you're not here today. And if you're in the world of maintenance, and asset management and reliability, that is the first place to go SMRP.org make that happen. We're gonna wrap it up on the other side. Stay tuned, we will be right back.
17:33
You're listening to the Industrial Talk Podcast Network.
17:43
All right, John Todd says gold in that data. If you're not in the data, analyzing and collection game, your competition is a guarantee. So find that trusted individual, that trusted company. Right there. John Todd, reach out to him. I can trust him. You can trust them. I highly trusted. He knows what he's talking about. And given the fact that he's been out and Goldstone in the middle of nowhere. Yeah, he's got street cred. Alright, gold in that data. Alright, once again, Industrial Talk is here for you. You industrial professionals amplify your voice. You need to do that. You you need to solve that problem and be able to help people solve their problems. That's why we are here. You go out to Industrial Talk. You're going to connect with John. But then also you're going to connect with me just click. Can you be talking to me? Let's have a chat, chat or two. All right, be bold, be brave, dare greatly hang out with John change the world and we're going to have another great conversation shortly. So don't go away.
