Session Information:

Reliability Centered Maintenance and Asset Connectivity

The ability to predict outcomes can prove to be a competitive advantage in any realm of life. Prediction, within the Maintenance community, is key to staying out in front of equipment failures. The importance of a clean source of data in order to identify trends and detect failure modes especially random failures is what keeps equipment available for use and reliable when in use.

The ability to collect data is an obstacle, especially in capital-intensive industries. The solution is an investment in a cloud-based operating system that, through sensors placed on specific components of equipment, will allow the transition of asset data into the ability to monitor and analyze that data that then is used to make inferences from.

These inferences then lead to the deployment of specific maintenance strategies.

Key Takeaways:

• The role that Reliability Centered Maintenance (RCM) plays within any industry is critical.
• RCM is a systematic (data-driven) and standardized approach to define and execute routine preventive and predictive maintenance activities in order to maximize company assets ROIC and capacity through improved equipment availability & reliability and reduced operating and capital costs.
• RCM relies on asset (equipment) data in order to make decisions as to what maintenance strategy to deploy such as preventive or predictive.
• Many organizations who utilize RCM are not realizing the full-potential of it due to a lack of clean data from the assets (equipment) they are committed to manage.
• Leveraging process mining to turn event data (equipment in a run state) into predictable outcomes drives many advantages for the organization such as staying out in front of equipment failure.

Session Transcript:

For excellence for global container terminals coming to us from Jersey City, New Jersey, we have Robert Cartier with us. It's a real pleasure to have, Robert, with us. Let me, let me give you a little bit of insight on Robert's tremendous background. He's an innovative, focused executive with experience across nine different industries.

He has a proven track record of over a billion dollars in cost reductions and generate revenue.

He's a developer of the machine order of analysis and author of the soon to be released book for phase approach to Competitive Advantage. So, Robert, I'm dying for some insights, maybe a little preview of what you're going to be covering on that book here. But first of all, thank you so much for taking your time to share your insight and expertise of our global audience today.

Josey, thank you for an amazing introduction. Always good to be back presenting with details. So I do appreciate that.

OK, well, thank you all for taking the time to attend this presentation. And I'm hoping that you all walk away with something from it but from my end I'm really hoping to make some connections within the toast and Networks civically, within a process mining to provide some, maybe lessons learned or even some guidance for me moving forward in my journey.

So, My executive summary here, you know, I want to state that, this is a hypothesis, as I've not been able to find any application of process mining.

With what we call reliability centered maintenance program as the architect of the RCM program within GE, transportation.

During the GE digital transformation error, process mining was only used for the analyzing and monitoring of product data such as locomotives, aircraft engines, and health care equipment and not the machines that actually made them.

So, after I walk you through kind of a high level introduction of what this thing we call reliability centered maintenance as, I'm going to introduce you to one of the strategies that fall under the RCM umbrella, which is predictive maintenance.

From that point, we're going to talk about the types of failures.

And discuss those a little bit. Just to get a better understanding and perspective of how predictive maintenance ties in with specific failure types.

From that point, we're gonna look at types of data leading into digital transformation.

Then, the differences between process, mining and process mapping will be defined from a very high level view, then finally, a case will be made for the application of process mapping, for Reliability Center means.

So reliability saunter maintenance is a systematic data driven and standardized approach across the company to define and execute routine preventive predictive maintenance activities.

Said another way, it's an asset management, focus with the objective of predicting equipment failure, through a focus on how the equipment actually fails, Which we will call the failure mode, the effects of the failure, which we call the failure effect, and the solution to the failure as to prevent re-occurrence.

Elements within the reliabilities under maintenance framework includes computerized maintenance management system, which we currently call C M M S Many versions of that out there, My GE days and my days and is still industry. We use an app with ... called maximal here on Global container terminals we actually use to we use what's called acid lies in Canada and dossier here in the US.

The second element is: corrective maintenance, followed by preventive maintenance, followed by predictive maintenance, spare parts management, and metrics, and auditing.

Reliability sunder maintenance: maximizes company's assets, return on invested capital, and capacity through improved equipment availability, reliability, and reduced operating and capital cost. So, you might be asking yourself, So, what's the difference between RCM in any traditional maintenance program?

The key thing as we move forward to, to understand, is that traditional maintenance programs do not solve to root causes of the failures, They do not identify failure modes, they do not focus on the risk of the failures. And they don't focus on predicting and preventing failures, but rather just the symptom.

So, kind of, uh, simplistic view of these types of maintenance: that, you know, we just looked through, pulling from the RCM elements in the previous slide.

The types of maintenance are presented here with a focus on, under what conditions they actually apply.

So for purposes of this presentation, we'll focus on the predictive maintenance strategy.

Predictive maintenance focuses on those not so obvious or not easily detectable issues, as you can see with random failures, not subject to wear, or what's called PM induced failures, which PM: acronym for preventive maintenance. So there's some cases where a too much preventive maintenance is occurring.

And it's actually creating some, some issues down the road that aren't always detectable.

As an example, if you think about it this way, getting your car serviced at the dealership would be an example of a time based preventive maintenance.

You know, it's time for your oil to be changed, It's time for your spark plugs to be changed, whereas oil analysis, taken from an F 16 fighter jet, which I was involved with my many years in the United States Air Force.

Is an example of condition based predictive maintenance where in that case, you know, you're looking at the oil analysis for detections of metal shavings or any type of debris that would lead you to take further, further corrective action.

So, let's hop into predictive maintenance a little bit, and talk about that.

So, defined as a comparison of trends of measured physical parameters against known engineering limits.

And the goal is the detection of the potential unhidden random failures, which I mentioned a short time ago, through the monitoring equipment, to determine the current condition and viewing trending analysis.

We're trying to stay out in front of actual failure of components.

And that's why we're taking the steps to use what's called condition based programs.

So, the property, it, this is planned. It's not unplanned.

So as an example, unplanned, a component fails on a machine. So now, you know that stuff's production that stops the process. You have to stop and react to it. That's unplanned. It comes at a very large cost to any organization.

This is planned, This is put into the work orders, there's time allotted for it, and thus, there's no interruption to process, and it's condition based monitoring. So it's not based off of time. As an example in my industry, you know, we discharge large vessel ships with cargo containers. We also load those containers back, we deal with very large equipment.

In the past, accounting has used the accounting useful life of the equipment, as far as the total cost of ownership, whereas, ideally, you want to go by, how many hours are on the equipment diverse, How old the equipment is.

Because, thermal though the terminal, depending on the volume of containers, the use of the equipment, and the environmental conditions, the run hours is a more accurate representation of what's really going on.

Then, under the predictive technologies, under predictive maintenance, we have five types of testing and inspection here.

Vibration analysis, this is use, obviously, to measure, detect any type of vibration within, you know, on a shaft and a motor gearbox. Infrared is used for scanning, hotspots on motors and electrical panels. Motor testing is used to detect any inefficiencies in the motor itself to include any type of rotational.

Um, no occurrence that's going on.

Ultrasonic testing is used, you know, as an example, detecting a burr in a metal bearing, An oil analysis is used to detect any type of metal or debris within.

The oil.

So condition based programs, which we're calling predictive maintenance, extends the service lives of equipment averting costly unplanned downtime and minimizing the cost of replacing those expensive machines or equipment.

So, types of failure, we have potential failure, So, a physical condition, which indicates that the failure process has started.

As an example, you detect metal particles in the oil, something has already occurred.

Hidden failure, random failure with just a little bit more difficult failure is not apparent until the function is actually attempted, then you realize that something had occurred.

Then finally, the worst-case, the functional value or this is the catastrophic failure, the inability to meet the specified performance standard.

This, this would be an example of bearing, just totally seizing and then operation of the unit, that that bearing was in no longer is operating.

So again, the goal of the use of predictive maintenance is the detection of potential.

In hidden failures, we want to use predictive maintenance to not encounter the functional failure. We want to stay out in front of it.

So that leads us into a great tool. Please don't be alarmed or too. Scared about this graph, it is pretty busy.

This graph is a lot to take in, so please bear with me.

The importance of understanding the types of failure, as we just discussed on the previous slide, is apparent with the use of what we call the P F curve.

The PF stands for potential for failure and the actual failure itself.

The benefit of using this tool is to determine how early a machine problem or misuse can be detected. So, how do I read this?

So, if you look at the P one changes in vibration, What this is stating, if left unchecked that between 1 and 9 months, that you would experience the functional failure, the F down at the lower right-hand side there.

And you're repeating that with P two.

You find some debris in the oil If unchecked and no corrective action, you'll get 1 to 6 months and then you can expect to see failure.

You notice that the P two to P three point, condition monitoring, IE, predictive maintenance, can be used to prevent the condition from progressing to the next P and eventually, to failure.

It's a great tool to use.

This is an example of one on an 1800 RPM bearing season, while working in, They still want history.

I use this to predict wall deterioration inside ingot, soaking pits.

And we were able to benefit from this type of information using this type of tall to a point where we were able to schedule our maintenance windows within days before we would start encountering issues with the soaking pits.

That then would lead to called the Issues and thus shutting our process down.

So, maintenance analytics.

When I was working for GE, one of my or the most impactful statements I heard from a company leader at that time was maintenance analytics drives world-class maintenance organizations.

So, you know, with analytics, you can begin to peel back the layers of the onion.

And I can't say that most maintenance groups that I've worked with don't really understand how to use data to detect trends and thus stay out in, front of failure's.

Tiina went on to say that there's questions to ask: what are we finding? Where is it happening?

and why is it happening?

And, again, this goes back into my, my comment, that most maintenance groups that have data don't necessarily know what to do with it, or, in some events, they don't have as much data that they would need to make some intelligent decisions.

two categories of data will break them down between lagging, leading indicators.

We have failure. data is captured through, let's just say, entered maintenance work order. So, a maintenance technician goes out. It works on a machine.

He comes back end, that machine had failed.

He came back in the, submit his work order, that gets loaded into a CMS, which I referenced a little earlier.

And, from that point, that data can be tracked for trends of failures on machines, but, at this point, the failure had already occurred, thus the lagging indicator.

And then predictor data, which would be the leading indicator and this is captured through what we'll refer to as digital transformation.

Once we view this data, a failure has not yet occurred and thus is a leading indicator, which is a predictor of the lagging indicator.

And you can see some examples of predictor data, which is the temperature pressure vibration, so on and so forth of machines. This data then would be maintained within data warehouse or data warehouses.

Interludes for decision making. So, as mentioned earlier, GE wasn't a digital transformation error during my time. There is a Lean six Sigma Master Black Belt.

GE began installing sensors on products such as locomotives, which I was involved with. That was my division.

Aircraft engines and healthcare equipment as a means for GE to monitor the health.

Other products for plan preventive maintenance purposes, which was a service that they then would provide.

So, relying on sensor data was critical in order for their products, not fulfill on the field. And that was a big no-no for any GE product.

So, pre dex, which is a GE cloud based operating system was introduced during this time and it's still being offered through GE Digital now, you know, I'm being told that it's not a great operating system and there's other better ones out there.

So, you know, maybe some of you can message me through LinkedIn and, you know, give me some recommendations of some better ones. But you can see in this example where on one of our newer at the time evo, locomotive Engines, 250 sensors sending over 150,000 data points per minute.

And so, in this data, would be analyzed to determine, you know, when preventive maintenance was, was due, and then start going through the scheduling process.

How close that locomotive was to a certain point in its route where that service could be carried out.

So, digital transformation. So, in order to transition from the analog to the digital, machines must be equipped with sensors instrumentation software, systems and tools, et cetera.

I will say that machines equipped with programmable logic control, will call PLC, contain sensors, but may stop short of having a capability for digital transformation.

I will notice this, or I have noticed this in order industries, such as still industry or the industry that I'm in, A lot of the equipment is older and order meaning older than 20 years old, And some of it has, PLC being controlled by PLC and some of it doesn't.

Um, you know, in my current company, we're in the phase now of identifying all machines that have PLC sensors, where they are located with respect to systems of systems and components. And then matching what those sensors are detecting to what we feel are actual needs, ultimately, will be from a monitoring and analyzing data perspective.

So, you know, Asking the right questions helps you better understand the needs for that transitioning to occur. So, you know, you just don't decide. I'm going to transition from analog to Digital, Now.

I'm in a digital transformation space without really understanding, no, why should we transform? Are we behind? What data do I care about? What goes in?

What goes out?

What are we collecting? And why are we?

So digital transformation steps. Transition from analog to digital, which I mentioned at my current company, Global container terminals were at that point right now within one terminal of our four terminals. using that as a pilot.

Creating a Digital Twin, oh, the machine.

No, Digital Twin is a software representation of a physical asset, if you will. System or process designed to detect, prevent, predict, and optimize the real-time analytics to deliver business value? Which, at the end of the day, that's what it's all about.

If you think about it, it's it's really defined by the outcomes.

It's trying to achieve such as reducing unplanned downtime, which I've mentioned already, which is an improvement in reliability of the machines.

The endgame is to transition from preventive maintenance, two predictive maintenance, and it's all about that early warning.

So an example of a Digital Twin would be a GE aircraft engines and locomotives based on the operational fleet data of components.

Pumps or compressors critical assets, turbines, or systems of those assets. So, a lot of people are familiar with seeing it as a three-d. model.

Collect data in data warehouses.

I will say that even though, that this comes in at number three, the creating of the Digital Twin and collecting data in data warehouses can be happening in parallel.

Analyze and visualize the data.

Learn through process mining, then finally, deliver value on the asset.

I apologize.

So, providing a different view of the process, now, we're talking about process mining a little bit.

You know, the original view derived from the engineering view, if you will, and lacks what may be actually occurring real time every time.

So influence the process through process of mining, know, some things here, more efficient, faster, more robust, more complete.

Process mining is based on data, and not assumptions, which, you know, in the past, from time to time we all get caught.

A little off guard, we're probably directionally correct but not precise working off of assumptions, it's complete as it uses all recorded processed data, including all exceptions and variance.

It's fast, as it's fully automated, and is repeated at any desired moment.

one great example would be how Amazon delivers packages in your garage if you're not home.

No, They probably asked themselves looking at the data.

Customers weren't happy when they couldn't get packages.

They looked at the data to say, Yeah, that's definitely a problem for us. So, how do we solve this and keep the customer safe and happy?

Then, when they made the change, they probably saw that the customers were more happy because more packages were being delivered.

So, great graphic. I really liked this. The first time I saw it, it really hit Home for me.

So, traditional process mapping off to your right.

You don't see kind of the end around better occurring within the process, because you're using how the process should theoretically, be working or just a small snapshot in time versus the end around that.

No happen quite often on the left side with process mind.

So, your deviation from the standardized process with maintenance can, no, can be not completing the preventive maintenance, or completing but not to a satisfactory level can also be the manner in which the machines are being operated to include.

The environmental conditions.

I've been part of and facilitated many process mapping sessions.

And at the upfront deep dive, data analysis isn't completed prior to the event, the mapping is only as good as the information that is offered by the participants.

So, as a note, you know, sometimes employees deviate from the standardized process and process mining, the benefit that you get from there, from the application is, it detects when that does occur.

So now, comparing process mining versus process mapping, keeping it no high level digitizing, which all the digital twins and process maps, doesn't provide any benefit as a standalone.

And, you know, I seen this in GE, specifically within the four walls of the factories, with the machines where we weren't leveraging process mapping. It was just occurring with our products.

So even though we had a lot of good data, or if there was a digital twin on the machine, that in itself did not really benefit anybody just as a standalone Process mining is leveraging that data across those machines, as an example.

Within the processes and learning, the new insights through that data, what's it telling us?

Process maps are ideal or snapshots of a process, whereas having streaming data will uncover the unforeseen issues and it's that real-time view, that you get that.

It's really impactful, and provide solutions that have data to support and model the investment.

So, currently, we're in the industry four. Speaking about predictive maintenance, which falls within this fourth industrial revolution, Wikipedia defines it as Industry four is the ongoing automation of manufacturing industrial practices.

So, organizational dynamics will begin to shift and evolve within industry for us, four, specifically, around the role of acid connectivity, IT Support, what, eventually, what it means in terms of qualifications and skill sets To be pursued in maintenance and asset management organizations that has digital twins for their assets and then on within the RCN journey.

So, RCM, leveraging process mining is great.

Um, Visual here, The purpose of the two types of process states that you see in the pictures is not for you to be able to read them, but for you to visually see, there is a difference. In my earlier example with GE and the use of predicts with their products.

What I am proposing in this presentation is the same overlay with factory machines or process machines or even equipment that I see here at Global container terminals out in the terminals, um, with these types of products, I would like to just to see that overlay. And, I've never seen that, in my career, yet.

So, process mining would detect changes through modeling.

A something is wrong, from the engineering documented standards, such as, an example, for pressure, temperature, vibration, wear, and those variables are very important to you, then those are the things, that, you want to focus and on.

The detected changes, then can be monitored through the use of predictive maintenance.

So, that modeling is pointing you to be more precise with where and how often you utilize the predictive technologies we talked about earlier.

And then the advantages of predictive maintenance is that it can detect the root cause of then, what is wrong.

Plan maintenance thing can be completed, thus avoiding component functional failure, unplanned downtime and the impact to operations and finally, customers.

And, finally, so, circling back around of what process mining can do for RCM.

Your process mining investment drives full maturity of RCN. As I'm seeing it Now, you know, someone mentioned yesterday, the process mining might be on its way out, so, we'll see how that works.

But, knowing where to apply predictive technologies, the type of technology and the frequency has traditionally been identified prior to process mapping and maintenance through failure mode effect analysis, where the systems of systems and components risks are identified.

Having the ability to be more objective based on data, not assumptions, more complete.

Using all recorded process data, including all exceptions and variance, and then fast, fully automated and repeated at any desired moment, through process mining not only advances the ...

program but becomes an organizational competitive advantage through the reduction of that unplanned downtime and its unfavorable financial impact.

So I thank you all for, for taking the time to allow me to present, and I look forward to taking any questions.

Terrific, Robert, thank you so much for bringing this perspective to us. I have questions that have come up here. And then the, I want to encourage the audience to keep asking questions, because we have even a little bit more time here to engage in the discussion with Robert and learn from his expertise.

So, Robert, the first question and maybe overall main theme that has emerged is just, uh, set the stage for us a little bit more. Some people are not familiar with what the company does and how you operate, so just give us kind of a dimension on kind of the types of operations the company has. You know, what it involves, you know, locally, globally. Just a bit of an overview about GCT, if you could, please.

Yes, yes, thank you for that question.

So GCT is what's called a Marine Terminal Operator.

So if you've watched movies, you've seen commercials on television.

You see the large ships coming into the port, and, you know, containers are removed. There, put back on. Those containers are either staged in the yard Or they're put on trains or they're put on trucks. We have a location in Staten Island, New York.

We have one in Jersey, City, Jersey, and then we have two and the Vancouver, British Columbia area.

The company has been around for quite some time. This is a very small industry.

Customers are very limited and the customers that you do have to sign long term agreements. So, there's not a whole lot of wiggle room for a lot of things there.

But where you can capitalize in this industry is with the maintenance, because it is capital intensive, and at the end of the day, we want to keep these these machines up and running. We want to keep them available for operations.

And when they do have them, we want to ensure that they're reliable.

They're very, very well in the operations themselves. I mean, how much of it is main engineering and maintenance versus, you know, other roles and just kind of a high level picture, in terms of the Personnel Division?

Yeah, so operation's represents the biggest component of the organization, but followed closely by maintenance.

So, there has to be a really good relationship between maintenance and operation again, maintenance is ensuring that the assets that operations need to do their job and thus service. The customers are there, and while they're available, they're operating as advertized.

That's great.

So, the, The, one of the other themes that it has emerge here is about the, the work, that the, In the maintenance area that you described. Which is, by the way, fascinating. It's a deeper dive, because we talk about technology, sometimes, and, and, often, and it's, and you have kind of this IT view of the world, right? And you and you have provided a real operational and maintenance view of the world, and how does it connect with that IT view of the world?

Which is really critical, because the business outcomes, you know, are ultimately what we want to impact in a positive way.

So, what has been the journey at GCT, with respect to implementing technology, in their maintenance area, Because, I have to assume that you have been doing, you know, of course, maintenance for decades, probably. But, but, the implementation of different levels of technology in your maintenance era, that has been an evolution there, as shown, what does that look like, and, and when the process mining become a thing?

Yeah, great question. In this industry, I'll say, even within GCT, prior to 2016, it was traditional maintenance that I mentioned in, in the, in the presentation.

Around 2016, we introduced Reliability Center means here.

So we started really focusing on, OK, instead of just changing the oil, let's, let's understand, why are we doing things, and how things fail, and the effects of them, and then how we can prevent it, from happening.

We went with predictive technology, we're using three of the five, but I mentioned we're using infrared, we're using oil analysis, and we're using vibration analysis.

We're at the point now where we're not in that digital transformation phase, we want to be.

And then finally, we want to apply process mapping, or mining to enable us to be a little bit more intelligent with our decisions, and be quicker to respond.

But right now, as I mentioned, we're at the point of looking at some of these older equipment and saying, OK, this large crane is controlled by programmable logic or control. And we know that there's X amount of sensors.

The things that we want to monitor through process mining.

Are these, are they represented by where the sensors are, right, And that's the phase we're going through.

Now, you know, understanding this is going to be an investment, not only in the transferring from analog to digital, but a little bit of an investment in the actual process mining itself. So, very early stages.

And, you know, anybody who's watching, please send me a message through LinkedIn. Again, I'd love to hear some lessons learned, or even some guidance. And your early experience with your early stage journey process.

They add that, that, that is, that is fantastic.

I think you said in your presentation that the digital transformation doesn't do anything. I mean, it's like if you're not born digital and you have to go for analog to digital, there's a, there's an evolution that you have to go through there to really harvest the fruits of digital transformation, right? And you're talking about the need to put sensors in place to start collecting data. And some of those things are not trivial. Because the investment and the disruption to existing operations can be significant.

So one of the questions that has come up is that do you prioritize, how do you look at your operations and say, OK, these are the areas where I can create the most value with digital transformation. Is this is this, you know, how how do you do the fatty recreation prioritization when you look at operations and maintenance and decide where technology is going to be tried Eve.

Well, first of all, I would say it starts with, you know, the board of directors, right there. There's an investment, and it starts at a high level with business intelligence platform overall.

And under that platform, there's many arms that come off, obviously, a process mining, digital transformation, so on and so forth, falls out from underneath of that. So, you know, a company has to have that commitment to that long-term investment. You know, once that happens, then, you know, when I look at John or Global container terminals, I look at, OK, where is the most value that we can get from, say, Process mining? It definitely would be in the operations and maintenance groups as compared to, you know, maybe a Finance or an HR.

Not that it can't be applied, but, um, yeah, we're at that stage now, with the BI platform being approved, the investments there. So now all this early work is going on.

And again, it's all about being, making better decisions, quicker decisions for the benefit of our shareholders and our customers.

That's excellent, Robert. And Robert, tell us a little bit about how you're engaging the people on the ground, right, The maintenance personnel, operation leaders and and and the people who are executing on a day-to-day basis. How are you engaging them on this digital?

I hate to call it a digital transformation, let's call it the digital journey for now, on this digital journey. And, because there are a couple of things, First of all, you're gonna get their ideas. You want to get their their input. But also, there is a potential fear that some of these technologies may affect their, their positions and their roles in their organization. So talk us a little bit about the cultural aspects and the human resources aspects of this type of journey for you.

Yeah. You hit the nail on the head with the fear.

You know, within this industry, it's, it's heavy on union workforce, and then your frontline, your company management that you have, you know, there's a little fear on their side.

So, you know, you have to manage that. I think on the company side, it's easier to manage it, right?

When you look at what the long term, um, no benefits of using something like this is, you know, as an example, they don't come into work and firefight as much. Right. Because we're going from reactive to preventive. Predictive, right. So, we're staying out in front of a lot of stuff, and nobody wants to come to work for 10 to 12 hours, and just be reactive all day.

Right, so it's the sink in a little bit, smarter, you know, I would think with the men and women who proudly serve in our union ranks between Canada and the US, you know, that's a little bit harder play on that side.

So, you know, that, that takes a little bit of work And maybe somebody who's watching this presentation that's spent time, and still industry, or the automotive industry.

Some of those older industries that traditionally have had, you know, men and women in the union ranks working, will find no Or maybe can realize how difficult that is. So, it's going to be a journey.

I don't have an answer right now of how, you know, when we get to that point, we're going to win that. But, you know, it's got to be a process.

Absolutely, absolutely.

The reality is, and the challenges off of a transformation. Now, I'm gonna get some specific ones here. I have Anna Martinez please, who is implementing process mining in her own operations. And then, and she comments that, first of all, great presentation.

And the, she says, a potential use case for maintenance could be process mining for maintenance orders on time performance, is one example. So, she cares about, and you may have touched on this already in your, in your presentation, but she's asking which areas are targeted for process mining in maintenance operations, you talk a little bit about the machines, and kinda looking at how they perform. Any other examples there of that, where you were pointing process manager?

No, great, I mean, it's these type of, you know, pieces of knowledge, the date that I get.

So, that was a great question, and, you know, I would say that anything, any aspect within the maintenance function is free game. I mean, if there's something that we can improve on and that was great with the maintenance orders. On time performance. Yeah, I wrote that down.

So, no, I wanted to keep it high level.

You know, as an example with GE, if we're doing this on products such as aircraft engines and locomotives, why can't we then just do it on the machines in your appointment, in our factories and in our terminals?

So, I'm sure there's a lot that drops down under that, but.

But, yeah, that's a good, once I always say anything, you know, is opening my eyes for the application.

Very good, very good. ... sense, a big shout out from Spain here. He says, great presentation, Robert. Thank you for showing the importance of the of maintenance for operations. So he's a, he's a kinda heart there, when it comes to, when it comes to maintenance and operations. This is a soft heart for it, for our maintenance operations. All right? So, why do we have here?

We have Lynn Fuller asks, What considerations should be made, or multiple streams of real-time data?

Um, I could interpret this in different ways.

Do you deal with that? Are you dealing with multiple streams of real-time data there, and then you have to make decisions about which streams of real-time data you're going to build.

You know, latching onto is your process mining tools at GCT? No, Not at this point. I mean the data that we're dealing with this data that's coming in through the CMM, S that I mentioned.

Um, But that's about it.

But, when I think back it, even my GE days, you know, where so much data came in, and it kind of touches into machine learning a little bit.

You know, there was just more data, then, than humans can actually, no, have time to too, you know, to monitor and analyze and make inferences from.

So kind of a tale of two cities between GCT in and in GE. But, you know, currently with GCT, we're not dealing with that problem. I would like to deal with that problem. But having data is, is, is key.

I hope I answered that question.

Know that, that's very, very good, and, as a matter of fact, I'm reading over, another question that he has, it's a nice follow up to that one, and, I think, gives a little more clarity. He says that his personal experience is, with its, with manual examination, measurements, and data collection, not so much. with, with this, actual, with this real-time data streams. And the, and his questions is that, What sources do you use for the data, are you using? And maybe this predict your data, is specifically when you're looking for predictive data. What are the sources? And his question is, Is mother equipment now, especially, like in the operations that you have? When you have equipment there is: Is it routinely delivered with data capture capabilities? And I would guess I'm going to pre FTSE that I assume that you're not working with brand new equipment That's kind of already built-in sensors.

I'm assuming you're retrofitting some old equipment to get the sensors in place so that you can then get the data streams. Is that correct?

Well, you know, we do have some new equipment.

The company has invested in millions, but the what is coming with is the program will the logical control, the PLC, um, so you know there is that capability a little bit easier compared to an older equipment to go from analog to digital. And why we care about variables such as pressure, temperature. No vibration is, again, those become predictors to something ultimately failing.

And the goal is we don't want something to fail because in this industry, and I've never seen this in any industry, but this one, we don't take time to have maintenance windows. So maintenance never has a window to do any maintenance on any equipment unless there's not a vessel parked here, you know?

Birth here where you know Operations isn't working or you know.

We have a vessel that was supposed to be here and didn't show up So It's it's very difficult in this industry for maintenance to even you know get on the assets that To fix them, But Again, this is that whole analog to digital Step one of the digital transformation phase, if you will, but older equipment no order equipment or newer equipment.

Yes, it's starting to come, But it still doesn't give you. I think everything they all.

It's not touching all the points you needed to touch Well said. Robert, I can speak all day about this subject, it's a real implementation of process of mining on on an analog world that that the that can benefit from from the digitalization. But it's but, again, it's a journey and it's great to see or evolution and maturity on this. and I wish you very much success. And thank you so much for taking the time to share your expertise in the middle of this journey with our global audience today.

Thank you, everybody. Thank you for the great questions, and please hit me up on LinkedIn. I'd love to hear some of your, your lessons learned and some of your guidance. Thank you, everyone.

Terrific.

Thank you very much, Robert.

That's Robert Cartier, vice-president of Operational Excellence, for GCT, talking to us about the importance of the real transformation and digital transformation that takes an analog heavy environments which are, believe it or not, still most businesses today. And so, this is great lessons off of what's going on in the world. Yesterday, for those of you who were with me with us yesterday, I was in Houston as a matter of fact. After our conference and I spend the rest of the day with with energy organizations in Houston yesterday and the end of the issues that he talks about are very are worth talking about doing maintenance scene 2000 heat exchangers eat heat exchanger waste to £20,000 OK and the and the How do you assess systems like that? And, you know, they don't have sensors on them that's telling you what the temperature is and those as different things at least the old equipment doesn't.

And it's a, and the decisions to digitalize some of those assets is not trivial, or significant investments and in the return on investment. And the business case sometimes is not straightforward.

So, the real challenges of, of, have the asset operations, and, in this case, operations and maintenance. So, again, connect with Robert, Go under the posting that we have for this conference. You can see, you can look up, my name is just appears on LinkedIn. And see the posting that we have, and the under that, you've gotta have comments.

You're going to have questions to speakers. I'm gonna put, you know, some summaries of the things that we did on each one of the days. So, go check it out and ask questions that were not answer, you know, during this session. Now, we're going to take a break now. When I come back at the top of the hour.

You're going to be with me and we're going to have a journey on the world of excellence and innovation acceleration.

I'm going to share with you the empirical evidence and experiences of over 30,000 global excellence and innovation leaders in more than 100 organizations in over 20 countries in the past three decades.

What does it look like when you have a great enduring performance in your organization? And, and, so, I want to give you more than that, I will meet you at the top of the hour again. And we'll talk about excellence and innovation, acceleration and unleashing the power of great people And organizations. Will talk about technology, we'll talk about ideas, will talk about methods, and we'll talk about people. And that what matters most when we are going through our digital transformations.

So, see you back at the top of the hour and for now, Enjoy your break!