IFS Assets CollABorative: Think Tank – Preventive, Predictive, Prescriptive: Evolving needs in EAM with Jon Mortensen, Global CTO EAM at IFS

Date of Meeting: 20 June 2024 10:00 AM US Eastern Standard Time

Jon Mortensen Presentation:

Slide: Preventative, Predictive, Prescriptive: evolving needs in EAM

• What I'm hoping to do is to lead us through some concepts and my ideas, they're not related necessarily to the product itself, but the idea here is to discuss the total concepts that are associated with it.

Slide: Who am |I?

• So just quickly, who am I? I'm the global chief technology officer for EAM here at IFS. I do everything EAM related. As we all know, IFS covers a lot of different products that do a lot of different things, but my complete focus is on EAM and what EAM does, how it sits in the market and what it does for our customers. So again, excited to be part of the Assets CollABorative and to get this moving.

Slide: Abstract

• The topic today is about the evolving nature of asset management, but with a focus on the maintenance strategies that we use and particular working our way towards moving our way away from preventative more to predictive and more to prescriptive. And as we go through that journey, there's a lot of information that we require as we go, that helps us understand what that process is.

Slide: Maintenance 3.0 v 4.0

• And when I think about these things, I personally think about maintenance 4.0 and maintenance 4.0 being part of the journey. We start life off 100 years ago with just doing a reactive maintenance and when things break fix them. But the reality is that that is not an effective way of doing things. What we have seen in the last 30 or 40 years as people moving to preventative maintenance. Having a schedule running and trying to fix it, but the process has been fairly simple. Had a problem, we've tried to fix it, we've had to get all the parts, we've tried to do inspections to make sure things are OK, set out all our repairs manually and do those repairs and try to record what caused those repairs in some way. I am more and more flabbergasted when I see great EAM solutions from multiple providers that enable people to collect fault codes, reason codes, symptom codes, all those sorts of things and not seeing them being used, because they're the sorts of details that people need to collect, to move from preventative maintenance and being saying, oh, this is why everything fails, to being able to use that information to help guide them in predictability, et cetera.
• But the idea with maintenance 4.0, the same as anything else really is trying to reduce the amount of downtime, and trying to increase the amount of uptime that's associated with our assets. And there's a lot of techniques for going about doing that. We're going to cover off on some of those today.

Slide: Asset Management Trend Report – December 2023

• But what I thought I'd do first is just share some information that we picked up last year. I launched I trained report which was an independent survey of around about 430 maintenance managers. The majority of those were actually based in Europe, with probably about 30% of them based in the US, but asking people just where they were on their maintenance journey. And it's always interesting, there is always a group of people who get involved in the sessions that I attended and do at different conferences who are still in that reactive phase, who have not gone anywhere, but usually have a system. So the system is just collecting information rather than being a management system. Most people, almost 2/3 are in that stage where they're doing preventative maintenance as well, and maybe a little bit of manual condition monitoring by doing manual inspections to see where they get. When they start getting to proper condition-based management, we're starting to see less and less people involved in that sort of activity. And why I find interesting about that is that to move from preventative to predictive, you need to understand what's taking place with your assets and if you're not doing a lot of condition-based assessment, that makes it very difficult to make that journey. And very, very small number of people who've got to that Predictive stage of maintenance. So, it's sort of an interesting journey.
• So, what I'm going to do just before I go into the next slide, I want to get a feel from the group that's joined us, but I have a small poll, which I'm going to launch on the teams app and you'll find in the chat area of the screen. And what I'd like to do if it comes up and hopefully it will, is get you just to quickly click on how you rate your own maturity. In in that curve and where you feel that you sit, I'm not collecting who's actually doing the responses. I'm not interested in individually where you were all at in this case, but just for the group's purpose of trying to get an idea of where the people on this call currently sit. How would you rate your maintenance maturity? Maintenance 1.0, Maintenance 2.0, Maintenance 3.0, Maintenance 4.0, Maintenance 5.0?

• So I've got the majority of people sitting in that maintenance 2.0 area where we're doing preventative maintenance. So, it's really great that we've got nobody sitting in maintenance 1.0, you're all customers of ours and I was hoping that you'd be somewhere else.
• Does anybody want to just share verbally about where they are in that journey and whether they're trying to get to as they cover that? I'm more than happy to just get some verbal feedback for the group?

Customer 1 – Feedback:

• I think this is really a combination of different strategy like maintenance 1.0. So we have some tactics on run to failure and 2.0, but of course, our ambition or vision depends on the type of asset and depends on the customer maturity because we are OEM. So we have a we have to understand the customer perspective and try to find the best customized maintenance strategy. So, I would not say that we are on maintenance 4.0 or completely on maintenance 2.0. It's a combination. So we need to find the right balance. Which one is really important? So when it comes to our customers, they are really prone to downtime, then we actually try to see and serve the maintenance strategy according to them. So I would like to understand if we have the similar situation in other group of customers, if they need to have a combination and a mix of strategy that would be really interesting to hear.

Customer 2 – Feedback:

• We run terminal operations, so waterfront terminal operations and container yard terminal operations, and so primarily mobile equipment. I sort of resonate with what you're saying. While we have a fair bit of proactive and certainly maintenance in control to paste depending on the age of the equipment and some sort of long rather expensive long lead items, we still find ourselves moving back and forth into a little bit of reactive, simply because we can't spend as much time as even as available to try to get in front of every preventative issue in some of our older equipment. So, I don't know that there's a certainly a significant bucket that we would fall in most of the time, certainly moving towards trying to get as much proactive as possible, but we definitely find ourselves moving within the purview of maybe into reactive the least amount of time, but certainly between in control, whereas we would have some sort of new, relatively new, certainly far more advanced technological equipment that we might even move into some of the smart, because it has more tools to help us be there. For us, it's a lot about resource management. Our core business is moving goods and cargo and so maintenance generally doesn't get given maybe the all of the resources that it would want, but I think that's how we swing a little bit between the pendulum and I guess the goal is to not swing too far one way and get stuck there and work towards keeping the spread of the pendulum swing a little bit shorter in between those buckets.

Customer 3 – Feedback:

• The Backbone for our maintenance strategy is the asset criticality. We assess which are the potential severity of a failure on a specific equipment with the likelihood that this occurs. And then based on that, then we try to decide which is the best approach. And as you said, sometimes the best approach is just run until it fails, because the consequences of having a failure or the cost of preventing this failure is much more expensive than just to fix it when it happens. So as soon as you choose a decision to run the equipment until it fails, that's fine. However, then for critical equipment, obviously we are aiming for reliability. That's the first priority that the equipment will not fail unexpectedly. And to do that, sometimes we are taking too much time-based maintenance approach, which means that we are consuming a lot of resources from maintenance, human resources, but also spare parts, taking out equipment from production to run the preventive maintenance routines, which in a way is giving the main goal that is that the equipment is not failing unexpectedly, but we are penalizing the availability of the equipment because while it's out of maintenance, is not able to produce. So, we would like to push some of this preventive maintenance greetings we have today to more condition at based maintenance routines and now we are exploring the possibilities obviously to implement much more advanced techniques on predicting maintenance like vibration analysis, ultrasound technologies, thermography and so on. So, we are at that at that stage now.

Slide: What is Asset Management?

• And then it's the maturity that comes with making those sorts of decisions. I've got a graphic coming up, which is about how to make those decisions and the sorts of questions we need to ask ourselves as we go on that. But it is an interesting philosophy from my point of view.
• I come from an ISO 55000 backgrounds and it really comes from the time that I spent in utilities back in the 1990s where I was introduced to the New Zealand Infrastructure Manual. And so I have this philosophy around asset management, price asset management, which has flowed through all the way to ISO 55000, that everything that we do really is a trade off between performance cost and risk. And that performance component is often for people, the critical assets and what they do, and what the value is of what we produce out of that. And then the risk of course is the business risk of something failing. And what the impact on the business is, so you usually do end up focusing on those critical assets which are high risk assets to make sure that they looked after at a different level than you do when you say that a piece of equipment that you've got, you can get redundancy out of in things like that because the risk goes down with redundancy, the cost can be ameliorated and so on and so forth. But to sort of get to that area, there's a number of philosophies that we often play with from an EAM perspective.

Slide: Asset Management System (Relationships)

• So, from a design perspective, when we start thinking about it, there are a number of philosophies that we have in asset management. There's the capital cost versus the OPEC operating cost that's associated things which is often known as asset investment planning. I don't know if everybody's heard the message, but we're in the process of acquiring copper leaf, which is a leading asset investment planning tool that helps you make those financial decisions of allocation of money. Asset performance management, which is really critical for bringing in the information that we need about the assets and operating. And as we get more and more smart assets, that plays more and more of a role in what we do and of course enterprise asset management, which is the core of where your data is stored, where you can bring all the information in and use that information to start making the critical decisions that we need to do in those areas.
• So, having those 3 concepts in place and thinking about performance versus risk versus cost. I'm interested to know, how many of you actually think about those three factors as you go through? When you think about how you design your maintenance strategy, do you consider the balance between performance, cost and risk?

• Four responses this time. The only person who said no was me. So it means everybody's thinking in that area and I only vote so I can get it off the screen. So that's actually a good thought. Any thoughts from you guys on the balancing of those three things? Is it a difficult thing to do? How do you go about it? What is it that you think about when you're thinking about those three factors when you make those sorts of decisions?

Customer 1 – Feedback:

• So basically, when it comes to cost, performance and risk, we consider the reliability part as the first part, and the criticality is a another point. And criticality we need to consider from the cost perspective. Our main business is selling filling machines and processing equipment and also providing services. So, safety criticality is there, but right now we are not actually touching those assets. We are actually focused on the production critical and operational critical assets and there we try to understand the reliability and production losses and downtime cost and then also the cost of preventive maintenance including the spare part because some spare parts are so expensive plus lead time. So there are many parameters that is associated. The current challenge is it's a time consuming process. Second, we still would say that our asset health monitoring, the condition monitoring part is still somehow disconnected. So we are looking for combination of RCM and asset health monitoring in combination, so that we will be able to do this cost risk and performance balance and to make an optimal decision in a more appropriate manner. So, there is one factor that we are really struggling with is a time consuming process. Number 2 is that our asset health monitoring is still disconnected with RCM.
• We have a very good way of analyzing. We have 4 pillars that we need to fix together. People, process, system, data. I think the people part is the most important because when you are dealing with thousands of thousands of asset, the change management is the key because then what we are talking here, not everybody understands the same way. So that's the first hiccup that we have to go and do the change management. That is people part. Then the process part, because it's also important how we are dealing the data. Here we are collecting the data which is needed, and that data could be sensor data, that could be historian data, that maintenance, and then we have another part. That's why we are working together with IFS on how to get that data in a way that we will be able to perform this ultimate APM approach or asset management approach.
• And then system wise of course we are investing in the technology. Currently we are still in, I would say yeah, the machine learning algorithm or AI application is still not connected on uncensored data. So, we are actually monitoring our vibration sensor and different sensors on critical equipment. But we are really not applying the advanced technology to do an automated update and automated detected anomaly and then we can actually understand the PF curve in a more appropriate manner. Where we have achieved the majority is the data part. So we are able to connect the different data sources at one place. That's the first hiccup that we managed to solve, but it's a long journey.

Customer 2 – Feedback:

• We recently installed IFS. I would say absolutely on the journey and hoping to leverage the tool to help us expedite the journey or make the journey move a lot quicker. I think the five items that you mentioned there are specifically people in process. We learned a lot about that during the implementation. And so yeah, I echo that. We're certainly on our journey and trying to get people aligned. We've had a lot of change in our business, a lot of retirements and a lot of newer maintenance management staff coming in. So, we had done a lot of work on process and then of course have to relook at processes. We were moving towards the system, so yeah, I echo what we're saying.

Slide: Balancing Reliability for Availability

• These sort of philosophies are good to understand and one of the things that was mentioned was the PF curve. And I have bastardized the PF curve, and I look at where we try to make maintenance decisions as we go through the failure process with those. Typically I look at things changing over time that we might go through a number of observed failures that have an impact on production but not a critical impact on production. But the production quality does start to deteriorate in many instances as they come through, and the equipment condition also deteriorates at the same time, so we get an increase in cost, both in operating costs and in people costs as well, which gets associated with that. But it's always a balance. And when you try to optimize this, you're always trying to look for that sweet spot of when we can do that. But until you get to that condition management phase and understand the condition of the asset, it's actually very difficult to actually go through that curve.

Slide: Decision logic for task selection

• Searching on critical assets to do things, and this is a graphic from reliability web who I read their book many, many years ago called Maintenance 4.0. I actually collect books on different things, so the other book I get used from is a Dutch book called VBM Value-Based Maintenance, which is actually very well worth the read as well. They've got some very interesting philosophies, but the mechanism of going through what should I do from a condition monitoring task? What should I be doing from a preventive maintenance task? Should I just let run to failure because that's an acceptable thing to do as an interesting philosophy. And when I go across here, I've got different types of things I look at. What are the safety consequences from a failure that I can actually see as opposed to what are the long term or knock on effects of failure which I get on the right hand side because I can't actually see those failures which are typically things that occur along the PF curve, but also what are the operational consequences from an economics point of view and the non operational consequences from an economic point of view and working my way down through what should I do as a basis for that. So, this is a structure and there's lots of different structures that are out there to do this sort of thing. I'm going to ask two questions. Who actually gets and looks at these different types of consequences.
• When you decide on the type of task you create for maintenance, which of the following failure consequences do you take into consideration? Economic operational consequences, Economic non-operational consequences, Safety impacts, Potential knock on failures.  You can hit multiple options on this one.

• So, most people look at from what I can see here, the economic and operational consequences and the safety impacts. But it would appear that we've got two responses also for the economic non-operational consequences. So that's an interesting balance. What do you look at when you're looking at those economic?
• Operational consequences and the safety impacts. What are the key things that come out when you're looking for those things?

Customer 1 – Feedback:

• As I said earlier that it really depends on the operational context. So, let's say talking about any specific spare part, then in some cases when spare part is very expensive, availability, ordering time. So there is a lot of inventory related aspect that we need to consider. Even though it's a non-operational context, but since it is very expensive then we have to consider that factor. So even though you are reliability might be high, but your availability is not there, like spare part availability, that's really an important factor. Then we also think about the downtime, the production losses. So those are the ideas. Those first, we need to understand. And then after the most important part, availability for doing preventive maintenance. So maybe your decision support can say that yeah, you have to perform this maintenance on a specific period of time, but are you available? Are your customers allowing you to perform this maintenance, so those are the factors that we need to consider for doing a optimization decision. I would say yeah we have some maturity achieved. We are able to do it, but the engagement from various stakeholders, such as reliability engineer, maintenance planner, customers operational window, that really affect us for making such decisions. So we have to find a window rather than a specific decision. So it's more like a spread then just a point that we perform, but we use data for making that decision.

Customer 3 – Feedback:

• I try to leave the things in a very pragmatic approach always. If your company in your strategy says that safety is first, and this is not just on a piece of paper, it's something that it really means something for you. We have to prioritize our preventive maintenance task to avoid accidents, and people can come back home safely every day. So in a way, this is a factor that we have to consider in our maintenance strategy for sure. And the second one, also being pragmatic. None of us are nongovernmental organizations. We are here to make money. So, there's always a balance between how much it costs to maintain the equipment, how much it could cost if I am not maintaining it. And based on that take the decision, also bearing in mind that the backbone of all this is the criticality of the failure and the consequences that this failure can have. Economically, safety or customer impact, or sometimes even reputation for the company, that if we appear on the newspapers that we have had a fire or we have had an environmental issue. A contaminating a river or a emitting out of the of the limit. This has also huge impact, so that's my point.

Slide: Establishing Risk

• This slide is based on risk, ISO 31000, which comes out of the old Australian, New Zealand Standard 4360, which is where I cut my teeth here to do those sorts of things. So, we should probably dive in at some stage on actual product and what we do on risk and what we're building on risk. But as I said, it's one of the things we're hoping to showcase at Unleashed this year. But that risk process is really important and I am actually interested because of that, what the first word is that people think of when it comes to risk assessments in designing maintenance strategy. So, this is my last question that I've got there. If people would like just to type in the one word that comes to their mind when they're thinking about risk as part of their maintenance strategy to pop that word into the into the poll.

• Somebody's come up with incidents. I've struck it with a ISO 31000 because that's what I think of. Downtime. I'm going to leave that up as we talk in case people come up with other words, but I'm also interested in in what sort of methodologies people are actually using when they do risk assessments and risk management itself. And I'm interested in this primarily because when I think about how we design and make our products go forward, this process becomes really important and understanding how people think about it. Downtime seems to be the big response, downtime and safety.
• I'm just going to wrap up here with some thoughts on then over overall process for moving from Predictive. We sort of gotten into that condition-based component and moving on to that final thought process which is umm Prescriptive. You need to get to the Predictive, before you can get to the prescriptive because prescriptive requires you to know that you're going to have a problem and what it is. And I mentioned that we were looking at building in FMECA, which we're going to talk a lot about it Unleashed this year. But it's an incredibly important part of the journey. Is knowing all of these sorts of things as you go through.

Slide: Maintenance 4.0

• So just two sets, this is just a conceptual diagram. I've been using it probably for about five or six years and I've only been at IFS for about 18 months, but this component feeling is that you actually need to understand what your failure modes are. You need to understand what your risks are. You need to have the information about your assets and those typical downtimes to get to the predictive. But what you also need when you start getting into the prescriptive is that FMECA analysis. What do I do when something fails, when I hit a particular failure mode? What is the cause of that? What is the effect of that? And then what do I do to ameliorate that? Is all part of that process and until you start building up those libraries, it's very difficult to get to that prescriptive maintenance component. We're looking at using AI to help drive the initiation of the FMECA process and to enable us to get to that next stage. But what is absolutely critical, and I mentioned there's right at the very beginning, is that connected worker in the feedback loop. If we predicted something has gone wrong or is going to go wrong, we found a window to do something about it, which is sort of the optimal time we can do it and we we're doing that because we understand the risks that are associated with it. And then we send somebody out to do something with what we thought was the problem. They have to be able to feed back to us what they actually found. They have to be able to feed back to us, did we get that right? Did that look like it was a problem that was coming up? Did I observe that when I was actually in the field, if I didn't observe that, what did I observe? Because I've got a circular process that I need. I need that feedback loop that's going to take place from that worker, that connected worker, to bring the information back so I can update my training models with that data to get better and better. So over time my models should be self-learning and should be coming up with more information to help me work out what the issue is that's sitting back there.

Questions / Answers / Feedback / Responses:

• Q: I came from oil and gas background as of now, construction projects, railways. I've used multiple systems. Many RP systems. In terms of your question about the operational side, it's 90% of this I would say, because if you consider the operational side of the maintenance, this will affect the safety from the negative or the positive side. I mean safety comes first. Definitely this is 100%, but the safety will be affected based on your operational. How you're going to manage your operational side of the equipment. We all know that the operation is a big resistance for the maintenance means, we have around 17 or 18 operational risks. The only thing they do is drilling. So, the operation and the last thing is they want to do is stopping the operation for the maintenance. I mean, we barely have a couple of windows during the month to perform our maintenance, right? So, I would ask a question about the operational risk assessment maintenance. How IFS is considering this when it comes to maintenance? If you go to the work order process and IFS the let's say the standard IFS work order execution process, I don't see anything related to risk assessments within the work order execution process. We have done something outside. I'm not saying outside the system, but it's customized, where we had to include the risk assessment and the maintenance risk assessment within the work order execution in IFS. So I'm not sure if I'm up to date with IFS, the new IFS, but I don't see the risk assessment is a part of the work order execution process. Keep in mind that when you talk about oil and gas operations, a lot of our maintenance are being pushed because of the operation. So, a lot of high critical maintenance, we cannot perform it on time because of the operation. I mean we are in a specific operation. We cannot stop our drilling process. Resources are not available. Material are not yet received. Whatever the reason. So, I would say assisting this risk and putting a mitigation in place, it should be a fundamental step of the work order execution process. Where I at least, I don't see it from my side. If you could answer and elaborate more of the risk assessment where this risk assessment is something we can see it in the world called process in IFS.
• Q: Can I just ask what version of IFS you're on?
• A: Currently we are using IFS 10 UPD 15. So we have customized the risk assessment. We have used the risk assessment under the safety module, the HSC module and have it as a part of our work order process. Whenever we need to deviate from the work ordered due date or planned finish date, or due date, whatever you want to call it. I mean, we have to raise manually at risk assessment and again it's a customized process.
• A: The risk matrices and things that we provide in the solution at moment sit within that HCM module and at the moment we're redesigning the FMECA and the risk processes that we have in the solution. We're hoping to showcase some of those things, including risk assessment at Unleashed this year. So, it's probably a very optimal time to start thinking about it, because risk, as you say, is absolutely critical for some of this decision making that we do.

• Q: My thoughts when we are thinking about maintenance, the maintenance procedure itself, it's one process that relates to the production process. The production process but instance if you have a schedule downtown, you need to see that in the production scheduling. Have you been talking or do you have any information regarding that part? The planning part over the production tool that we have in the system. There is information, but is that extended somehow or officialized or better in the coming versions?
• A: One of the things that we have been working very hard on is using our optimization engine to interact with the production components of our ERP system to enable the scheduling of windows for maintenance as part of the production cycle. So taking information both from the maintenance side of things and from the production side of things and doing a prioritization using our optimization engine that goes with that. I thought we'd already released that as part of Cloud, but it may be that I'm reading something on the road map, but the concept there is to try to get that balance between production and with maintenance. Historically, I have actually sat down with a production manager and a maintenance manager to ask them how they dealt with this. They were using a whiteboard, but before I could finish my conversation with them, they actually ended up in a fist fight over it. So, I actually left and let them sort their differences out without me in the room. That was a very embarrassing episode in my life. But I I'm not sure if that answers the question or give some direction to what you're asking.

• Q: I'm not sure if this is related to this subject, but I would love to discuss this with you guys. Just for your information, it's coming within our organization that a part of the succession plan for our employees is to have the people certified and based on their positions, they have to be certified of what they are doing. Let's talk about the case. I have some people in my team. They work as support for the maintenance, users, so I would like to know where we can send our employees to a classroom and they get certified and EAM module in IFS. I'm not talking about online classes, I'm talking about classrooms where they can be conducted in Europe, Dubai, whatever the location. If it's something available, and I mean, I would like any feedback from you, is this something available from IFS or from IFS partners? Anything like this? A classroom we send the employees to get this training a couple of weeks and get certification exam or whatever the assessment comes after the training.
• A: Typically, we do that through online, not as in class format, where people turn up to the class. I will take that as a question on notice and go back to the training department and actually and inquire into whether any of our partners do that. I don't know of any off the top of my head, but as a takeaway I'll go through and find out for you and then circle back with you. That's probably the only answer I can give you just at the moment on that front.

IFS Cloud Asset Management – Unleashed 2024

• IFS unleashed. We've got 3 breakout sessions that we're going to be doing some focus on. These are actually interesting because the first two are based on the reliability process and what we're doing moving forward, which is in the area of asset performance management and the last one actually addresses a question that somebody else had on here about maintenance planning and scheduling. And so there will be a session which actually starts driving into that in more detail as it relates to maintenance on the shop floor and that relationship, I'm pretty sure will come out between production as well as part of that.

Reliability Leadership Experience

• We've partnered with reliability web. It's an organization being around for about 25 years in North America to provide reliability certification and libraries typically around the United States. We've just ran the first one of these in Amsterdam this week, which was very successful, but we've got 3 coming up, but the other three are all based in North America, but we are at the moment working with reliability web tools. So see if we can get a session operational as a side at Unleashed for reliability and for Americans, these actually count towards continual professional development as well. And this journey that we go on from moving up the maturity curve gets more and more reliability focused. And as we get to that predictive and prescriptive stage, it becomes more and more critical as part of that process.
• July 11 in Washington DC. July 19 in Chicago IL and September 18 in Houston TX.
• If it's of interest to you, please contact jon.mortensen@ifs.com

Next Meeting:  September 2024 TBD

IFS Assets CollABorative - Tech Talk Session: How IFS is innovating with AI

If you are an IFS Customer and you would like to join the CollABoratives, please click here to fill out the form.