00:00 Ryan: Hi everyone. Welcome to Masterminds in Maintenance, a podcast for those with new ideas in maintenance and reliability. I’m your host, Ryan, I’m the CEO and founder of UpKeep. Each week I’ll be meeting with a guest who has an idea for how to shake things up in the maintenance and reliability industry. Sometimes the idea failed, sometimes it made their business more successful, and sometimes their idea revolutionized an entire industry. Today I have with me Rob Kalwarowsky, an experienced reliability engineer and an event speaker, a coach and creator of Rob’s Reliability Project which offers the maintenance and reliability community with audio-visual information for how to spread the word about reliability. I’ve been his guest on his podcast previously, and now I’m so excited to welcome you, Rob, to our show this time, welcome!
00:51 Rob Kalwarowsky: Yeah, thanks for having me, Ryan. I’m excited to chop it up with you today.
00:54 Ryan: Alright. [chuckle] So let’s get started, let’s jump right into it. Could you help and start us off maybe by sharing a little bit more about you, yourself, your background, and really how you got started into the reliability space?
01:10 RK: [chuckle] Yeah, so I am a mechanical engineer and graduate from MIT. I have an undergrad in economics as well, and actually I started my career working as an economist for about about a year.
01:26 Ryan: Wow. [chuckle]
01:28 RK: Yeah. And then I kept looking for kind of mechanical engineering jobs and I found one at Teck Resources in Western Canada, a big mining company, and so when I moved out there, they put me in a reliability engineering role with Jeff Smith, who was actually my manager. Now he teaches for reliability web, and so that’s really where I started working in reliability. And then since then, I’ve worked in reliability consulting, and now I’m working at Enbridge as an asset manager, obviously, in oil and gas space.
02:09 Ryan: That’s huge. So it sounds like you started your career actually as an economist, and then you’ve got yourself into the reliability world. Is there anything that you would take away from working in being an economist to how that relates to anything in the reliability space?
02:28 RK: Actually, there’s a fair amount, like I think as reliability people, a lot of what we talk about or a lot of the projects that we do it’s good to quantify benefits, it’s also good to quantify cost, and some of those benefits they’re not necessarily hard benefits, like “Hey, you’re gonna save $10,000 cash.” Sometimes it’s safety benefits, it’s environmental benefits, it’s even stuff like corporate reputation. And so having that kind of insight in how that all works plus understand and cash flowing in our present value, all that stuff, it’s really helpful.
03:07 Ryan: Absolutely, that’s super interesting, that’s extremely interesting, Rob, where you approach reliability with the economist hat on as well and you take a lot of those learnings and bring it over to what you’re doing today. That’s really, really cool. Kind of transitioning a little bit, people, I would say, have different definitions of reliability and it kind of depends on what space you’re in, and what you’ve done in the past and what you hope to do in the future. To you Rob, what does reliability mean to you?
03:41 RK: [chuckle] Well, so I have my reliability mug. I don’t know if you can see it. Maybe I can move it close.
03:47 Ryan: Yeah, okay.
03:50 RK: So it’s a mug with the actual equation for what reliability is.
03:55 Ryan: Alright.
03:56 RK: So that’s obviously reliability. If you’re looking at it from a math stand point, it’s essentially the probability that your equipment will succeed over this given period of time. So you have to understand what’s the function, how does it functionally fail, what’s the operating environment, and then what’s the duration? So that’s the traditional, if you Google reliability on Wikipedia, that’s what that is.
04:20 Ryan: Yep.
04:21 RK: To me it’s… To me I’m more, I guess, I’m more long on the asset management side of it and I really just think it’s how do we add value to our company and how do we make our equipment run more effectively so we can do that? A lot of what I talk about in reliability like some people will call that continuous improvement or they’ll call it like, whatever technique you use or whatever technique you really… Whatever background you come from, at the end of the day, we wanna make more money, we wanna run safer operation, we wanna have less bottlenecks, we wanna be more environmentally friendly like all these things are important, and if you’re using RCM or you’re using Kaizen or Six Sigma, I don’t really care.
05:14 Ryan: Yeah, it’s always about improving, being better, and I think it’s actually very interesting that you bring up this point of there’s so many different terminologies whether it’s process improvement, whether it’s you call it reliability, you can call it maintenance. Essentially, you’re right, the bottom line is we are trying to improve, we are trying to be better, we’re trying to make equipment, our entire business run more effectively and efficiently. And I also feel that maintenance reliability gets a bad rep for this idea of maintaining status quo. But really it’s kind of the opposite of that. We’re improving our processes, we’re improving our day-to-day operations so that we can be better. It’s not just about maintaining status quo.
06:02 RK: Yeah, and that’s kind of a misconception and I also think a lot of us we come from companies that as a reliability engineer you’re reporting to the maintenance manager, and I think that that’s where you get pulled into the day-to-day. The maintenance stuff is your, “This has failed today, I gotta bring it back to running.” Well, reliability, we’re in a business of what’s tomorrow, what’s next year? And I think that it’s not just a maintenance thing. If we’re getting into the design of the plant, we’re getting into our storeroom operations, what type of equipment we’re gonna buy to maintain the function that we need, like all these things, how do we even operate it? That’s another big thing that we should be touching on. These are not maintenance functions but I would consider it, as a reliability person, I would consider it good things to do as a reliability engineer.
07:01 Ryan: Absolutely. How do we begin to change this narrative for people to move away from thinking about reliability from just this idea of firefighting and maintaining status quo? How do we change the perception?
07:15 RK: I think it’s just stuff like this. I think it’s conversations that need to be had. I think that it’s communication and I find engineers, as a whole, myself included, and which is funny ’cause I run a podcast and I do all this stuff but we’re notoriously bad communicators.
07:37 RK: And I think that that’s part of our bottleneck, right, is we fail to communicate effectively to the shop floor and to our upper management and so that’s why we don’t implement stuff the best we can.
07:53 Ryan: Yeah, absolutely. Alright, so you’ve done a lot of work in this space in reliability, what’s the future look like for this industry in your opinion?
08:06 RK: Yeah, I love this question because I ask it on my show as well. I just really think it’s an interesting thing to talk to people about and see where their heads at. And for me, obviously, I think artificial intelligence machine learning is gonna come more deep into the space. It’s kind of on its crest right now, but I don’t think people are really getting the value they can out of it on that’s a separate debate for a separate day. I also think that you’re gonna start seeing things pop up like augmented reality, virtual reality, where they can be used to help people in the field with their job, whether they can see the work orders on their safety glasses or they can hear instructions in their ear, those types of things, they’re sort of coming, but right now we’re still in the infancy of what that would look like, and it’s just gonna build out in and 10, 20 years from now, I think it’s gonna be really cool.
09:14 Ryan: Yeah, and I guess to second that question, where do you feel like we are today? Maybe if you could share some stories, some insights, some examples that you’ve seen. Where have you seen the best reliability programs today and what did those look like?
09:31 RK: Yeah, the best reliability programs I’ve seen, the one thing that kind of identifies them to me is they have a champion, an internal champion. And that champion doesn’t necessarily have to be the reliability engineer or the reliability manager, it could be in some cases it was like a lubricant specialist or just like a mechanic who was really passionate, in some cases, it was a planner. Those are really huge. And then the other ones are… Those are kind of on the, I guess, the single plant type sites. And the other good ones I’ve seen have been… They’ve really been large mining companies where they’ve put in a lot of infrastructure, they’ve hired a lot of reliability engineers, and they’ve kind of set up their organization so that reliability has its own department, and they’re not reporting to maintenance managers, they’re reporting to the reliability manager who has enough authority and enough foresight to understand, “Hey, we’re looking at the next five years or 10 years,” type of thing.
10:38 Ryan: Yeah, absolutely. And I’m curious, some of the innovation that you’ve seen in the space today, what are you most excited by?
10:47 RK: Yeah, I really like machine learning as a concept. I actually sort of self-taught myself how to build my own machine learning models in Microsoft Azure, and I think that that’s kind of the next step, and you’re starting to see it with some of the service providers is they’re starting to understand that the reliability engineer should be the ones that are building their own models, not just this data science guy who comes from working at JP Morgan who’s gonna just show up and be like, “Give me all your data, I’ll run some models and we’ll be good.” And so that’s kind of what I’m excited about. And that’s more in the short term than the long term look. I see that being big in… It could be even big now, but it’s probably big and probably the next two years or so.
11:38 Ryan: Yeah. So maybe just elaborate on that a little bit more. Does that mean like taking in vibration data, running machine learning algorithms on that vibration data to predict failure? Is that what you’re seeing?
11:54 RK: There’s that. There’s also… I mean, a lot of these platforms, they have the ability to really take in any type of data that you give it. So a lot of the wins actually that we’ve seen, well, I haven’t seen them personally but I’ve talked about with experts, have been actually on the production optimization side, so a lot of plants, they currently have a SCADA system that has a lot of data that helps run their plant, and so by running those and training them on a machine learning algorithm and having the operator kind of… It just can make recommendations to the operator who’s allowed to still play with it themselves, they can optimize production. I’ve heard two, two-and-a-half percent which is significant over a year, yeah.
12:40 Ryan: That’s huge. And what I also love about this, too, is one is that you self-taught how to take in all this data using Microsoft Azure. And what I also hear from you, too, Rob, is that it’s this idea of, it doesn’t have to be the reliability engineer, it could be the plant manager, it could be a lubrication specialist that really becomes a champion. What I’m hearing from you is that anyone can do this. It’s really about taking that first step.
13:12 RK: Yeah, I mean, I’m not recommending everybody go out and learn machine learning. What I do recommend is that you should understand how it works and how to apply it at your site. And I think that’s a lot easier than people think. When we just talk about it, it seems complicated, “Oh, I need a PhD. This guy, he went to MIT, he’s a genius.” Whatever. I’m not a genius. I went on Coursera and edX, I took a course from Microsoft. And you can kinda get started. And I just recommend play with it. It’s no different than when you have your Alexa or your Google Assistant at your house. Play with it and you’ll find that there’ll be uses for it, and then just let your mind be open for creativity.
14:03 Ryan: Yeah, absolutely. Very cool. And then, just thinking forward a little bit, what do you think are gonna be some of the most impactful innovations in this space for the future?
14:15 RK: Well, so here I’m gonna give you the the most boring, unsexy answer ever.
14:20 Ryan: [laughter] I’m ready.
14:23 RK: So, we talked a lot about the machine learning, augmented reality, virtual reality. It’s not gonna be any of that. It’s gonna be getting back to the grass roots and really going out on the shop floor and talking to your people. And I was actually talking to my colleague about that yesterday. I was like, “What I want to do is I wanna go out to one of the sites and I wanna find the oldest guy and I wanna buy him a box of donuts and I wanna sit down, and I wanna talk to him about this specific piece of equipment and understand what actually he does in terms of maintenance, what he looks for, why he would decide to replace it or versus repair it? What types of repairs are common? What does a rebolt look like?” All these questions that I can’t answer sitting at head office in wherever I am, right?
15:15 Ryan: Yeah. That’s awesome. That means a lot to me, too. When you’re mentioning… A lot of companies think that, you’re right, “I’m gonna just augment our team. I’m gonna hire a data scientist to come in and look at all of our data, and they’re gonna have all the answers.” I agree with you, Rob, that it’s not gonna be the data scientist, it’s gonna be the people that know how this equipment works, that it’s gonna be the people that have all of this historical knowledge of what happened, when it happened, what makes it tick, and what makes it turn? Those are gonna be the people that will help drive a lot of the innovation in this space. And I wouldn’t say that it’s necessarily a data scientist skilled with looking at millions and potentially billions of records.
16:05 RK: Yeah, and I think that the reason for that… The way I’ve got it sorted in my head is, I feel like a lot of these data scientists, they’ve come from finance or these different types of backgrounds where they’re dealing with, I wanna say, unconstrained reality. So the financial markets, there’s no box around it, there’s no physics; it’s just like anything could happen. They’re talking about behavior, where we deal with things that are physics-constrained. A pump can only fail 10, 15 different ways, but there is a cap to what that number is. And especially if we go out and we talk to that old guy, he’s probably seeing 12 out of those 15 different ways. And so if we can capture that information, we’re gonna be way further ahead than waiting for 50 years to see all that data.
17:02 Ryan: That’s a very, very interesting point. You’re absolutely right, that we’re dealing with the physical world, and when you talk about finance and economics, there’s an infinite number of ways that things could change. But physical equipment, you’re right, 10, 15 different ways. Rob’s Reliability Project, the thing that you had up, you guys produce an incredible amount of very useful content. Curious, where do you find some of the most common questions to be in this community? What are people most curious about?
17:40 RK: I think the most… Maybe it’s not a question, but the most struggle comes from people who are trying to implement their reliability initiatives, and then the people who are trying to change culture. The technical information, if you wanna learn how to read vibration analysis, or you wanna learn how to replace a bearing, these things are very well-documented, you can take courses from numerous different companies, right? And so that’s not really what people struggle with, it’s the communication, it’s the… “How do I trigger change?” Which is a question that I’ve really been thinking about a lot lately. And it’s that human aspects, and I think there’s a lot to learn about that. I still don’t know much about it, but I do think that the big thing, communication, and then the second tip I would have for that is the communication part, you gotta speak to your audience not for you. And I think we often forget that, as reliability people, we’ll talk about, “Oh, you know the vibration. The FFT was this.” I even had one the other day where I was talking about RCM and I forgot that my audience didn’t know what RCM was.
19:00 RK: And so, we gotta understand, what does our audience understand? Like a manager who has an MBA and has just come in from accounting school, they’re not gonna know what the shop floor guy knows. And so when we’re talking to those different audiences, we have to be careful with how we juxtapose that conversation.
19:22 Ryan: Absolutely. Any tips for folks that are trying to make this cultural change at their organization? Any tips for them on how to make that? What have you learned so far?
19:36 RK: Yeah, so my big tips for that would be get out on the shop floor, put your boots on, get out there. My favorite thing to do as a consultant was to eat lunch and have coffee with the mechanics, ’cause you just shut up, just sit there and they’ll tell you everything you need to know, you don’t even have to say anything. And then, you can put it in a fancy report and charge lots of money, right?
20:04 Ryan: Oh, man.
20:07 RK: But that’s really what it is. If you wanna get back to reliability, you wanna actually improve reliability, you have to affect the behavior of the people who physically interact with the equipment, whether that’s the operator, the maintenance guy, the guy who lubes it, the guy who buys it, those people, they’re the ones that you have to change, and so you have to understand what are they dealing with and then you can’t make their job harder. That’s another thing that we forget as reliability people.
20:39 Ryan: Yeah, absolutely. We’re talking a lot about cultural shifts and cultural changes within an organization. I would say the next question to ask on that is how do you track it, how do you measure it, and how do you see the importance of having metrics KPIs within a business, what do you look at and how would you explain the importance of having clear KPIs and metrics for reliability and maintenance team?
21:12 RK: Yeah, I’m huge on KPIs. Now, when I say that, I also recognize that people do tweak their behavior or even some people tweak the KPIs themselves. What I would like to say is I love KPIs if nobody else could see them except for me. And that’s because I’m a data-driven guy. I like to dive into it, I like to troubleshoot, and I like to understand what’s going on like at the equipment level, and I think that that’s what the true reason for KPIs is. It’s not to bonus a guy, it’s not to yell at another guy, it’s about how to understand what’s happening in my organization, and then be able to drill down far enough to understand what I can do about it. And so what you’ll see a lot in organizations is KPIs, you’ll have competing KPIs. So you’ll have something where it’s like, let’s say you’re bonused off of store room one, where it’s like you’re bonused off service levels or it’s like the percentage of time I show up and say, “Hey Ryan, I need this part. What percentage of time do you have it for me?” But then I’m also, maybe I’m bonused off the opposite of that. You see that in maintenance. So you see availability, they’re bonused off availability, but they’re also bonused off like PM completion, so they’re kind of opposites. So, you do need opposites, but you have to be careful that departments are competing against each other.
22:55 Ryan: Yeah, because then that goes back to this idea of culture within the company and the organization.
23:02 RK: Yeah, and then you’ll see the next question you wanna ask if you’re implementing KPIs is what behavior would this drive that I might not wanna happen? So I’ve seen some sites where they had… Here’s a prime example for one: I went to a site and they had a scheduled compliance metric. And so they were trying to… And so what they did was the planner would issue a work order for a filter change and have it be 12 hours. So the guy would spend 30 minutes do the filter change, and then do 11 and half hours of reactive work, but they were hitting 100% scheduled compliance. And it’s like, “That doesn’t help us do anything.” Yeah, you hit your numbers and it’s green at the end of the month, but you have to really get into those.
23:55 Ryan: Yeah. Any tips, advice, guidance on choosing the right KPIs? What have you seen work really well?
24:04 RK: Yeah, I think you have to have balance in KPIs again. So 100% or zero should never be the optimal amount for a KPI because they’ll hit 100% or zero regardless, it’s not gonna be, but it’s probably not what you want. The other thing I talk about a lot with KPIs is reading versus lagging. So it’s a big focus on the SMRP example, but it’s important because the example I always use is, let’s say, I wanna lose 10 pounds. Well, if I just look at my leg and KPI of how much weight I lost, if I go to the gym today and I eat steamed broccoli and roasted chicken breast, and I weigh myself at the end of the day, maybe I’ll lose half a pound, but it’s probably mostly water weight, and so it’s not gonna really help me drive the behavior I want, where if you’re using a leading indicator which is maybe like time spent at the gym or healthy meals consumed, those types of leading KPIs, they’re gonna really get you to the end result and it also almost rewards you for doing a good job on the way.
25:17 Ryan: Yeah, absolutely. Cool, that was extremely informative. Switching a little bit, biggest mistakes you see people making in maintenance and reliability?
25:31 RK: Yeah, so, biggest mistakes, I’ve actually done this early in my career. I think the biggest mistake is doing reliability at your desk. And I think that it’s back to that sort of machine learning technology that type of stuff is when I came into a reliability space, I wasn’t… I had never been really in on a shop floor, I had never really talked to a lot of the guys, and I was easily able to get into the data, do some optimizations, do some failure predictions and pull out pretty useful information, but if it doesn’t make it to the shop floor, if it doesn’t change how you’re running your equipment, how you’re maintaining your equipment, how you’re purchasing your equipment, it really doesn’t do anything like, yeah, you learned something, it’s fun, but you didn’t change reliability. And so I think as a junior person out there, if you’re listening, go out to your maintenance shop and just find people who know what they’re talking about and just follow them around, and just help them and ask questions.
26:44 Ryan: Absolutely, and I think this really resonates with a lot of folks where at the end of the day, it’s really the people, you can only do so much by yourself sitting at a desk. Kind of on this topic, what we also hear very often is that maintenance and reliability commonly goes unnoticed. One of the biggest things that we wanna do here at UpKeep is really surface the what we call unsung heroes, the folks that are maintaining our equipment making them extremely reliable, because oftentimes you don’t really hear about that side of the coin unless something really does go wrong. So I guess the question here, can you share a story of how some of the the individuals you’ve featured through your project has made an impact on our world or at their business?
27:37 RK: Yeah, so the one I wanna give you is actually LaWayne Smith. So he works at the University of Kansas Medical Center and he’s been in reliability, I think now it’s been about nine or 10 months, and he kinda came into the role. They had five maintenance zones that were each doing their own thing, and he’s got them all to sort of buy-in, he’s got them all, he wrote and got approved, actually he got it approved the other day, his strategic asset management policy you gotta push through. And then he’s also implemented some predictive maintenance technologies at his site. So in 10 months, definitely in less than a year, he’s really come a long way, and I mean, who can’t get behind that hospital, right?
28:27 Ryan: I love it, that’s really awesome. And again, what my goal is here is to really surface these stories because again, you don’t hear too much about people in reliability making a big impact because the impact that they’ve made a lot of the times go unnoticed. Their job, our job here is to be unnoticed. [chuckle]
28:51 RK: That’s right, and it’s something we’ve talked about before, but it’s the ideal scenario is we never have a failure and the plant just runs perfectly, right?
29:03 Ryan: Exactly.
29:03 RK: And if that’s the case, it’s almost like a referee, nobody knows you were there.
29:11 Ryan: [laughter] Yeah, everyone plays by the rules, right? What’s something you wish more people knew about with regards to maintenance reliability?
29:22 RK: Yeah, so I’m gonna take a kind of a different approach here. So we’ve talked a lot about the technology, we’ve also talked about the culture change and I think that there’s kind of those two camps in reliability right now. And there’s a third camp that we don’t talk a lot about and that third camp is sort of the math camp. That sounds fun. [chuckle]
29:43 Ryan: Alright, tell me more.
29:45 RK: But yeah, its those… There’s a lot of, if people are listening, the American Society for Quality offers the CRE exam, Certified Reliability Engineer, and it really dives into the mathematics of reliability whether that’s the LIBOR analysis, that’s the Poisson Distribution, that’s all this different stuff you can do, the simulations, the Monte Carlos, all that stuff and it’s… You kind of don’t see that too much in reliability. Like I see people who do the maintenance side and I see people who do the math side, they never really interact. And I think that there is opportunity there for using that to find significant savings and, like I’ve used it in the past and I think that it’s something that people need to learn more about.
30:38 Ryan: Awesome, very cool. When you say it I think, “Oh, you’re right. This is how we conduct Weibo analysis.” But you’re right, it doesn’t come up every single day which is very, very interesting.
30:54 RK: I don’t know if it should come up every day. I just think that it’s gotta be a tool in your toolbox. You can’t only do root cause analysis and RCM, I think you need to be able to understand the impacts of those root cause analysis from a mathematical point like what’s the availability of my plant gonna be, that’s important to know.
31:15 Ryan: Absolutely. Rob, where do you go to continue learning more and getting better with regards to maintenance reliability? Where do you go?
31:24 RK: Well, I talk to people like you.
31:28 Ryan: [laughter] I appreciate that. And again, I feel like this is a small community. We are all learning from each other. Yeah, and I really appreciate all that you have done for the community. I think you’ve propelled it quite some bit.
31:44 RK: Well, I try and that to me, one of the reasons I started the podcast was to… I have a hard time reading reliability books and to have conversations like this, like I enjoy having conversations like this, and so it’s great to just have people on the show who are experts at whatever they do, whether that’s root cause analysis, that’s RCM, that’s AI, and just ask them like, What do I wanna know? And so that’s really how I’d been learning. Now, also, I listen to three, four hours of podcast a day. So if you’re talking about learning other stuff like recently, I’ve been really getting into real estate investing, so I like to branch out every once in a while. [chuckle]
32:36 Ryan: That’s awesome. If there’s one big take away from me in our conversation today, and I think one big takeaway for all of our listeners is really like getting out on the shop floor, going out and talking to people that deal with these problems every single day, go out and talk to the experts in this industry, in this space of whatever problem you’re trying to solve, that’s where you’re gonna learn the most, and I think that’s a great takeaway for everyone.
33:02 RK: It absolutely is. And one thing you’ll find, if you’re listening, don’t be shy, you can reach out, we won’t bite. We’ll give our best answer right. And so that’s one thing that with my email list has been really nice that people I say hit me back, hit reply, I’m here. And so I’ve started having those conversations and those… And it’s great. I got an email yesterday from, I think, it was from Chile and so I’m helping him out with some oil analysis questions, so it’s great to have those conversations and don’t be shy.
33:37 Ryan: That’s awesome. So Rob, if people do want to connect with you and learn more about Rob’s Reliability Project, where can they go, where can they find you, where can they reach out?
33:48 RK: [chuckle] Well, if you’re on LinkedIn, you can’t not find me. But yeah, just robsreliability.com. If you wanna find the email list, if you wanna find some of the top 10 podcasts are there. If you’re listening to this as an audio podcast, it’ll be wherever you consume podcasts. So I have Rob’s Reliability Project it’s available on 10 different platforms, iTunes, Stitcher, Anchor, Google podcast wherever. There’s also if you like the short audio tip if you’re not listening for 45 minutes, I put out a daily one minute tip, it’s called Rob’s reliability tip of the day, it’s available on 11 different platforms, so the same iTunes, Stitcher, Google, and then it’s also an Alexa Flash Briefing. So if you wanna start your day off strong.
34:39 Ryan: I love it, that’s awesome.
34:41 RK: Yeah, and then for the video picture content that stuff just follow Rob’s Reliability Project on LinkedIn. Also, you’ll see the memes which have been fun lately.
34:53 Ryan: [laughter] I love the memes Rob, I love it.
34:57 RK: I think a lot of it… I was a little hesitant about putting it out first, I was like I’m not sure how it’s gonna go, but I think people really like them. I like making them.
35:06 Ryan: I love it. Awesome, well thanks again, Rob, for joining us. It’s been a pleasure, and thank you to all of our listeners for tuning in to today’s Masterminds in Maintenance.
35:17 RK: Thanks for having me, Ryan.
35:19 Ryan: Alright, take care.