Suppose you need to spend $5 Million dollars on gearbox maintenance this year. How would you spend it? Stephen Steen, Director of the Industrial Internet Services with discuss options using predictive health maintenance and how Sentient Science’s “life extension” models allow owner/operators to understand what components need to be replaced, when, and where. Owner/operators use prognostics to know what turbines are failing and where, what components in the gearbox are failing and when, how to budget, and how to manage both inventory and supply systems.
Stephen Steen also shows a demonstration of Sentient Science’s DigitalClone Live SaaS used to create budgeting reports for component and gearbox replacement over a 6 year period.
Natalie: Thank you all for joining us today. I just want to give everyone a few more minutes and then we’ll get started with our presentation. Looks like we have a few people joining us last minute here, so just standby and we will start in just a few.All right, it looks like we have the majority of everybody online at this point. Welcome back, everybody. I see a lot of familiar faces. It’s great to have you attend our webinar session again. And to those of you who are new, welcome. We do have these webinars once a month, so you are welcome back and if you missed any, they are all hosted in on our website.
So, today we are going to be talking about $5 million Gear Box maintenance budgeting. So, that seems like a lot of money but, believe it or not, it can go very quickly and we want to take you through the steps of how we can spend $5 million. So, for those of you who don’t know me, my name is Natalie Hill. I am the Marketing Program Manager here and I do see a lot of familiar faces and was able to meet many of you in person. So, welcome and thank you for coming.
My information is here on the screen. Please feel free to reach out to me at any point in time. I just want to take you through our webinar series. So, like I said earlier, we do do one webinar per month, three per quarter and we follow this schedule here on this screen. So, we always start with a deep dive about our technology, and then we go into a life extension application and, today, we will bring you to the industrial Internet. So, we’ll be talking about a higher level view of how you forecast your budgeting for gear box maintenance.
All of our webinars are on our website. They are under video library, on one of the taps in the right hand corner, so if you are unable to attend any of them, you are always welcome to download the recording as well as the slides. Again, please feel free to reach out to me and I will be happy to send you to any of those webinars in particular.
So, I just want to explain a little bit about this webinar. It’s a little different than our webinars in the past. We always do encourage participation from the audience and we normally ask you guys to hold your questions until the end. We are going to change that up a little bit today. Please input your questions throughout the whole session and we will make sure to interact with Steven Stein, our presenter today, and bring up those questions throughout the session. So, we want to hear your feedback in what we will be discussing what the audience thinks about our presentation today during the presentation instead of waiting until the end.
Without further ado, we are going to go ahead and get started and I would like to introduce you to Steven Stein. Many of you know him as well. I see a lot of familiar faces, so Steven is the head of Industrial Internet Services here at Sentient Science and he’s really our true expert, so I’m going to pass it over to him now.
Steven: Thanks, Natalie. Great to have everybody on the call again. I see a lot of people we’ve been working with the past year, as well as some new names. Just want to double check if you guys can all hear us on your end, so if you want to just send a quick note and the question just to confirm that and make sure that the audio is working.
All right, great, thanks. So, just an introduction. We’ve been doing this webinar series for quite a while and we’ve also been working with a lot of customers, a lot of wind market, in terms of owner operators, and [inaudible 00:05:13], and similar areas. Through the past year, we got to do a lot of cool discussions on just really how can we change the way we look at how you spend your money on gearboxes and everyone budgets a certain amount of money and really the question always came down to “How are we going to maximize that gearbox spend?”
So, we thought it would be cool if we just took a lot of the discussions that we had over the past year, a lot of the decisions that were made, and a lot of different points of information and generalize it just a little bit, so it’s not too specific about each individual company, but see some of the decision making that has been happening over the past year, utilizing some more of a prognostic approach as opposed to maybe a planned version of that.
So what we thought it would be cool is to just say, “Hey, what if you had a $5 million spend, maybe not represented whole wind turbine owner/operator and the entirety of the gear box, maybe representing a smaller group of that. If we had the $5 million spend on maintenance this year, what would you do with it?” That’s a general question, is how can you approach maybe utilizing that money a little bit differently to maximize, in the end, your cost of energy?
So, some of the questions that you have when you walk into that is you are trying to ensure your energy bill better. Really, that’s how everyone makes money with wind turbines and how you return on that investment. So, how do you really optimize a bit of that? What does that look like? How can I balance the spending for the immediate gearbox availability and address that problem, as well as the long term [inaudible 00:06:56] value. Is there’s anything you can do today at a little bit lower cost to then reduce your cost in the future? How can you balance that to really the most out of the dollars that you are spending?
One of the big questions is what are you going to pay for the gearbox and is there any way that you can get better pricing on gearbox replacement inventory and what that would look like? Really, not to break a bank then, spending above that [inaudible 00:07:22] ceiling. So, if you have a limited cash that you are going to spend that year, what can you do to optimize that so you can get the most out of it but not go over it and trigger anything beyond that?
Then second, other piece of that is you’re integrating that with financial managers as part of the operation side and, typically, the historic owner data was projection, know what the future problems are but maybe not representative out into the future. By that, I mean, I think the industry, we were pretty good at doing the next year and how many machines that look like, but really what could you do to manage that note two, three, four, five years out onto the future and what that looks like?
Now, really, the question comes down to whoever is managing that. What do you do with that? What does that look like? So, we’ll just address some of those questions, again, using an example region, in this case, and I will go through a little bit more of that in detail. But what I want to make sure we do is, I’ll make this an interactive as possible. There’s a questions box as part of this that you guys can kind of send any comments or notes as I go through this. If you disagree with me on something, feel free to throw it out there and I can explain a little bit more on maybe why this type of customer would make this decision, again, using the assumptions of their financial goals, taking into account their fleet, their failure rates, that sort of thing.
So, feel free to challenge me on anything, I will try to address them as I’m going through the slides, if you want to throw those to the questions box. All right? So, over the past year, again, we’ve really been focusing on, if you have followed a lot of the series and a lot of the work that we’ve been doing, the switch from going from more of a PPM. I’m sure everyone’s very familiar with the Plan Preventive Maintenance and moving towards more of a PHM of predictive health management.
There’re a lot of tools that are out there. We’ve talked about those in the past as well and so those would be some of the diagnostics capabilities that give you a little bit earlier warnings, so that you can reduce maybe enough tower placement and get a little bit better hit on those as opposed to full gearbox replacement. But again, focusing on what is a true prognostics solution, bring a little bit more of which is the planning horizon. So, when we talked about a couple of the questions that were asked, “How do you optimize a short term versus the long terms health of your fleet? Is there something you can do, if you knew your planning horizon in the future longer than a year, or so down to sites and individual turbine level, will that change the way you make decisions today?”
Then we talked a little bit about the diagnostic versus prognostics again, detecting failures that are already there versus predicting when they are going to happen in the future, and again, extending that horizon to you knowing what’s going to happen maybe even years in advance instead of days to months. Then, we’ve had a lot of discussion within these planning pieces that went around, “How do you use the idea of risk through prognostics and have a discussion around that so you can, again, change the way your approach what that looks like?”
So, as part of that, we’ve viewed a bit of a technology and, in this case, we’re re not really talking about how to DigitalClone Live or Sentient’s technology, but just imagine if you had a tool that could predict when and how these metric components are going to fail is really the core of this discussion, but, again, a review of the tool that we used to do this discussion. We are really looking at trying to predict the initiation of the failure within the material itself, so we are taking into account a lot of the operational designs, supplier, variabilities within that system to understand what that risk looks over time and, if you can catch that earlier within the turbine life cycle and you make some decisions, “How can you really extend that into the future?”
So, taking it from this original, if you did nothing and just upgrade it as it is today, and move and extend it along that dotted line out to the future, what would that effect of life be? So, if you have a tool that is capable of doing this, this an example of how Sentient does this on major components and really the tool we’ve been working with most of our customers to answer these types of questions, “How can you do that earlier and more accurately to improve some of these decisions?”
So, who’s all using these prognostics and how is it being applied? Here’s just a quick look. It’s across multiple, different industries, not just wind and I just want to make sure that we point that out. The context that we are talking about today would be gearbox replacement on wind turbines with that type of focus on what that would look like and how those decisions were made. So, you have to make assumptions that reflect what a wind turbine operator would do, but if you take just these assumptions to different other arenas, you can obviously use the same type of data and look at an approach that would be much different.
So, I did have one question that I would answer as part of this next assumptions, so if I take a look at an example that we are going to use today, and we are going to go through a couple of steps of some of the decisions to optimize that. The example we are going to use is a region, maybe with eight to 10 sites. A majority of that region would definitely be in this example, one major brand, there would be a couple of other brands within a couple of those scattered sites, but in general, maybe we are making assumptions and that there’s one major provider for this, maybe something around 600 gearboxes or more total for that fleet. We are making the assumption that most of it is off warranty and that there’s some sort of CVM within this mix.
Let’s take a look at that, as well as this particular customer, since they are off warranty, they are little bit longer in life. They’re seeing a bit of an increased number in gearboxes failures as part of this. And because of that, there’s a strong focus and pressure to maintain or reduce the [inaudible 00:13:46] apex. So, the financial goals of this customer really is to see what they can do to reduce that [inaudible 00:13:53] spend, so that’s a financial goal that’s very important on the type of decisions that you would make. Not just this year but into the future and that we use Digichrome life in this case as a prognostics system to do some of this work. So, we wanted to jump in real quick and do a poll question to help characterize what you guys are doing today versus the example we are going to give here. So, Natalie, you want to go ahead?
Natalie: Great, thank you, Steven. So, before we get started into that six steps that we have here on gearbox maintenance, we wanted to ask the audience a question on what your current budget for ONM is on your gearboxes? We have a variety of answers here, if you guys just want to take a few seconds here to choose one and, if it doesn’t apply, just let us know.
Steven: So one thing, the point I think is really interesting is we’ve seen a lot of different expected failure rates within different fleets and it definitely didn’t show the same budget amount as with what a lot of the customers seeing, and so I think this is interesting insight for the group, if we compare maybe with the industry failure rates look like versus what’s actually being budgeted.
Natalie: So, the results are now showing. It looks like a lot of people do not specifically budget for these gearboxes and then we have some scattered answers throughout. Hopefully, we’ll take you through some steps and you’ll encourage your companies to be budgeting for your gearboxes then. I will pass it back to Steven now.
Steven: Yeah, so I think that’s actually pretty surprising that we have majority of those companies that aren’t specifically budgeting for that. We know that the industry in general is seeing these failures and a lot of discussions we are going to have today is, “What if you were budgeting for that and if you budget was $5 million?” So, maybe this would help take some of the insight we’ve gained with some of our customers, and apply it a bit to the companies that maybe aren’t taking to look at budgeting for these different failures and what that looks like.
So, we’ll be glad to talk about that a little bit more, but we did have one question come in. We did make some assumptions here, you know, an eight to ten site region itself with 600 machines, but we know that a lot of the owner/operators out there today, some of them don’t even have that many, so the question really came back is does this scale up or down? Can we do this on smaller amount of machines versus a large amount of machines and do the recommendations fit? Obviously, recommendations are based off of the goals of this owner/operator, so a bit of it has to do with the size and what they want to get to but, in general, a lot the recommendations for optimizing short and long term failures, that sort of thing, definitely fit.
Then, as part of that, really being able to see how many years forward what that looks like with prognostics. What does that look like? So, definitely when we are looking through this, if you are a little bit smaller, if you are little bit larger, some of this is going to scale pretty well and some of the assumptions that we’re making, we sort of picked a medium or a large to medium size as part of this example.
So, if we take a look at that, we are going to start spending the $5 million. We will just take a look at maybe the first step. In this case, what we are going to decide to do is we need to have a little bit more information about the fleet before we start planning what that $5 million looks like. So, if there’s a PHM powered tool that can give you an identification, again, we are using our product visual Digichrome life for this, but give you an idea of what gearboxes look like from a failure rate and over the next three years.
So, not just the number of machines per year and just for the next year, but the next three years and then looking at where and then what machines? So, we’re trying to identify what machines, what the root causes of those failures would look like and any other issues there. So, definitely part of step one would be to get a tool that allows you to, some of our customers had some of that already and then some of the other customers didn’t have anything in place. So, in this case, we are going to assume they didn’t have this but that was part of the initial spend to get that working.
One good clarification question that came in which is what is considered a failure? So, in this case, we’re going to assume a full gearbox replacement as a failure. I will give you an example of this. If the component was going to be replaced uptower, there would be a way to expand the life. You have to spend money and there will part of the gearbox spend would be there, but if you have to fully replace that gearbox, you can’t do anything else to the point where you can’t use anymore. So that’s what we are going to consider a failure in this case. Thanks for the question.
So, after you get those numbers, what does that look like, what are you really focusing on? One of the first things we looked at, in this example for this wind turbine, this may not be true for some other wind turbines out there, but there’s a lot of return investment for derating, and so what you could do is then initially, “How can we get that failure rate down for this particular customer with their price of energy, examples that we used in this as well as some of the actions that you could take.” Really looking at derating as being a good option, so this case, of the X number of machines, 60 of them saw immediate value of derating to get those failures out and reduce that original hit.
So, that really gives the ability then to react and start planning for when they did go down. So, you have a little bit less number of them failing, and then you have some oil changes that we could do on some of the machines, so you can get ready for an uptower replacement and then improve on that schedule and what that look liked. So, there are some quick hits that you could do on this and, in our software, what we are looking to do is we do a return on investment trade off. So, we take a look at the machine, its operating history, we run that through our models and then you can do some life extension trade off of options to see then how that would impact life itself.
So, you could go in here and you could say, “What if I were to derate this or this? What if I replace the oil, what does that look like?” The kinds of things that would include not replacing physical components on and the gearbox in this case, or the whole gearbox itself, so that failure, the point where you can no longer operate the gearbox, would happen. Push that back, in some cases, a couple of years. So, there’s some immediate payback there as the first part of spending, so you’re going to have some lost energy you’ve got to account for in the budget, you are going to have the oil change and some other things to take into account as part of that.
Step three, again, using a prognostic approach and how you would focus on that is you need to go out and do some confirmation of what is actually happening in the field. So, if you use a PHM tool, you probably want to correlate that to an actual maybe [inaudible 00:21:23] scope of particular components. If you take an example machine which has some issues and we look at being able to identify what components is predicted to have an issue, focusing on that, being able to go out and take a look to confirm that, allows you then to start planning, “How does the actual field failures line up with the predictions on the machines that are predicted to be in, almost failure today, right, majors pitting, majors falling or maybe even a gear tooth missing.” So, you really want to spend a little bit of money there to confirm that and then start to plan based off of that and then recorrelate some of those models as a second step.
So, looking at part of the fourth step, we would then take the pieces and the action we took from two and three, so now we know what machines are looking to fail. We took some actions, maybe, to lower the amount of failures that are going to happen year to year and then we know which ones we’re focused on. We also took a look at and identified the actual state of those machines out in the field.
Now, the next step for this situation to really low down is how can we change the full gearbox failures, and move them to an uptower replacement? So, if we take again this example, a set of machines and we approximately about 20 of those can be converted to an uptower replacement for this year’s part of that budget. So, what that’s going to do is it’s going to take some of those machines from failure this year, push them out maybe three to ten years, depending on what that refresh of that component allows.
So, as part of that step four, you are really taking and pushing a lot of those failures. You are spending some money today on whatever that uptower replacement costs, but you are using that wisely in terms of identifying which machines can get the most value from that. Then again, without a component by component prognostics look beyond just the bore scope, it’s going to be hard to know how much longer life you are going to get. So, in our tools, we’re specifically were focused on doing a component prediction, so you get an idea, “If I replace this component, how much more life do I get out of this?” That will really help you to maximize you RY. Which 20 machines are you going to spend your money on is you can only spend it on 20 machines out of the 600 plus machines?
So, if you went ahead and did that and identified what those look like, you push that forward. So, regarding the uptower replacement components, I think Natalie will have another poll question. Just to get an idea of the attendees today, how many do actually do uptower components and what that looks like.
Natalie: Great Steven, thank you. I’m going to bring us to the next poll question and our poll question for the audience today is how many uptower component replacements will your company make this year? So, 20 plus uptower component replacements per 100 unit, 10 to 20 uptower component replacements per 100 unit, five to 10 uptower component replacements per 100 units or up to five uptower component replacement per 100 units?
Again, if you guys don’t budget for that, there’s also an option on the poll question. So, I will give you a few seconds here to finish answering. It looks like we are about half way through the answer, so I will then close the poll.
All right, I’m going to share the results now. So, it looks like the majority, again, don’t budget for uptower component replacements, and maybe after running through the rest of this presentation, that will be a different answer but, other than that, up to five uptower component replacements per 100 unit looks like the next most voted answer there. Steven, I will pass it back over to you to take them through the rest of the steps.
Steven: Great, thanks, Natalie. Great, that was interesting. It looks like there’s a bit more to do more to focus on uptower replacement as part of the budget. I’m going to assume here that you can do some uptower replacements to replace those gearboxes that you aren’t budgeting for, so if you are budgeting for those fixes, but really focusing on some of that. I did have a good question, how many actually participated in the poll? We have about half of the participants on here, so 15 to 20 people participated in the poll. That’s a give or take there.
All right, so moving forward, we talked about the uptower replacements. One of the big interesting pieces of this too is, if you have a tool, you don’t want to just replace the components uptower with what you had in there before, right. So, one of the big, important pieces here is to understand that if you can put in a component in there, spend a little bit more money on what that looks like, in terms of the original component that is out there, is it valuable to get a “upgraded component?” For example, for bearings in this case that we’re going to use. There’s a high speed bearing that can be replaced uptower, a lot of systems on the market and a lot of replacements on the market.
What really is going to give you the return investment? So, we talked about before that derating trade-off that we use on our products. Same thing is true, we use some of the trade off to understand a little bit better what the value is, the actual return investment made by an upgraded bearing. So, in this case, we identified eight of those that would actually need to be an upgraded unit.
So, if you look on the left, we have an example of different OEMs and their failure rates across the fleet and what that looks like. Then we said, “What if you replace some of those with different bearings? What does that do to the life of those machines?” So, on the right hand side you can see the flattening out of this failure rates and extension of life there.
So, one of the important things too again is, in the future, how can I spend a little money now to make sure I don’t have that problem in the future, really trying to balance the immediate payback that we did in maybe step two, where we derated some machines, did some oil changes and other activities, versus here where you are investing in a new component to make it last longer. So, we find this very, very important, so if you can take that gearbox, refresh it and maybe may be now it will last to the end of the life, to the end of the 20 years. And what does that really look like? And so we find this very important step as part of this.
Following that on maybe to the next step is now that you have done all these changes, right, is there a way to optimize the energy output across this? So there are quite a bit of upgrading scenarios that are out there today. I think that everyone is pretty aware of them. There’s Fuji, there’s control changes from an OEM, there are third party control changes, there’s upgraded controls themselves, physical replacement of the electro equipment, and upgrade some of the other pieces and hardware that are on the machine. If you look at all those things out there, is there anything you can do to optimize the energy output maybe again, looking at another round of derating or looking at another round of upgrading on some of the machines that you’ve now optimized against.
So looking at the energy plan, how can you then run another return on investment study after you’ve made these changes to close the loop and then re-optimize the way that fleet is operating to either increase energy output by making, again, investment into some hardware, some control, some upgrades. And so this could be one way to recoup some of the spend that you have there or further improve the lifing of those machines out into the future, so reducing what those look like.
So, I think that’s a very important step here after you’ve done all the hardware changes for the gearbox and extent the life of some of them. How can you reduce some of the work we did in steps two and three to increase the life of those machines?
One thing we didn’t want to leave out, if you have a $5 million spend and all the actions that we took, what that look like, you definitely want to have some amount of money to cover head count and operating cost outside of that. So if we’re assuming that there is some extra cost involved with some of the activities that we are doing and staffing available to do that, it’s either internal to the company or its third party ISP doing some of that work or some of the contract work, we want to make sure we don’t leave out maybe some impact head count that we have to cover as part of the operating cost.
Lastly, most importantly, save a little money, go on a vacation on an island. So Natalie, I think a poll question here. I think this is the most important one which would be, “Where would you go on your vacation, now that you have spent your $5 million wisely?”
Natalie: Thanks, Steven. So let me ask you guys what your favorite island vacation is? And I have a few options here and maybe we are missing some, but is it the Mexican beaches, the Caribbean Islands, the Mediterranean, Bora Bora or Fiji? So, I’ll give you guys a few seconds. I wish there was an answer for all of them, because that’s what I would be choosing.
All right, I’m going to close the poll and share the results. The most popular island is St. John, and I haven’t been able to visit it but our CFO, that was one of his favorite islands, so I’ve heard great, wonderful things and saw amazing pictures. So, I’m just going to pass it over to Steven for a summary.
Steven: Thanks, Natalie. So one thing to note here, you guys, more of you voted on the vacation than you did on the actual stuff related to your job. So, I thought that was interesting because, obviously, I take more vacations than focusing on spending money on gear boxes. So, just to wrap it up a little bit. We took those different steps and then we focused on what is the cost look like? We attached some rough estimates on cost to those different steps to toll up to $5 million and so we thought we would leave this here just for a second or two. This is all available, again, on our website afterwards, if you want to dig a little deeper.
But what would have cost to do each of these different steps? So, again, you are refocusing on upgrading some derating, losing some energy a little bit there, oil changes, what does it cost for some of that? These are example of an average for certain types of machines and certain types of actions that will happen there, totaling up. So, I just want to see how that got ranked again because we are doing return investment focus, so money was involved in the decision making as part of this. So, we wanted to make sure you have that available here on this slide.
Really, in the end, what can we learn from this, right? If we look at the approach maybe that we work with our customers within the wind space the past year or so and some of the learning we have from that. What’s interesting is we know diagnostics and vibration monitoring. We absolutely could improve the ability to do uptower component repairs, but even taking it further, if you had, on top of that not just a CPM system, but a PHM system, planning for that, not just for this year but multiyears.
So, if you are able to hit a little bit more and we noticed there’s a little bit more of you that plan for uptower replacements versus full gearbox replacements. How could you plan that not just for this year, but next year and the year after out into the future which, again, could help you plan for a big dramatic reduction in the cost from a full gearbox replacements to just the uptower replacements as much as possible?
So, again, that’s the big swing there with the PHM is that you can do that on a multiyear basis of just a single year or maybe estimating out into the future using some historical failure rates. Past that, when you do have to do full gearbox replacements, which does happens, one important piece is that we can take a look at our PHM through a 10 year horizon. So, that should help then give you the capability of pricing inventory, supplier contracts, what does that look like to make sure that you are getting the best deals you can.
So, now that you have a really good confident idea of what that looks like from a failure standpoint onto the 10 year horizon, what does that look like? And related to this one, one good question is what are we noticing average gearbox life, since they are not lasting 20 years? If we take an average look, not the worst, not the best, seven to 12 years seems to be the ball park, depending on the type of install, what the machine is and that sort of thing. That’s one of the questions that just popped up.
Really, if you kind of know the average life, what does that really look for the 10 year horizon across your sites and how do you plan for that? But not just doing it from a fleet standpoint and number of years, but also what sites are going to have those issues, since a lot of those operate as a separate LLC.
Fourth one that we are taking a look at here. It’s now possible to maintain the fleet using prognostics again, going full planning instead of just PPM and diagnostics. Both of those have led to great improvements to how we do maintenance, but adding the PHM part adds another layer of reliable multiyear budgeting planning that we didn’t have before, as we use this part of this example here. Then, further prognostics help extend the RU of components. We are really focused on how we could make machines last longer because we know earlier, then we will start to then balance the immediate fixes versus the uptower long term fixes to reduce failure rates over time.
So, that really gives better input to planning and reducing your own budget, not just for this year but then out into the future, so you can continue to hit those targets and look at lower ones. So again, kind of the bullet point there is better balance between short term repairs spend and long terms asset value. All right, Natalie, I think we are going to roll out here and see if there’re any more questions that pop.
Natalie: Thank you, Steven I do have some further questions. I’m just going to start, unless you’re seeing something that you want to go ahead and field first.
Steven: You go ahead.
Natalie: All right, thank you. So, I have a question here from the audience that says, “How does your modeling work for hydraulic pumps?”
Steven: So hydraulic pumps, we do do pumps. We’ve done them in aerospace and some other areas. So, the same way it works within a gearbox. You would really be focused on the mechanical pieces of the pump itself are going to break over time and what does that look like for a lifing standpoint.
Natalie: Great, thank you, Steven. “How are your predictions validated?”
Steven: So, we’ve done quite a bit of validation. We did 10 years of RND before we brought this to the market. So, we have with quite a bit of universities, about $25 million of [inaudible 00:37:26] RND grants from different entities within the United States. We work with lots of different our customers today but in RND projects to prove this out. So, we have 20 plus validations of very clear results and capabilities and seeing what that looks like.
Natalie: Thanks, Steven. So, I see that there are a lot of more questions, and again, I’ll encourage everybody to input them in this as we are finishing up this session. I have another one. It says, “Do your recommendations scale up or down depending on your budget? My budget is smaller.”
Steven: Yes, we kind of talked about that earlier and that one really comes down to, in most cases, they scale pretty well because you are looking at the number of sites, number of machines, but because you’re smaller, you might have different goals, so then your return investment model maybe a little bit different in what that looks like. So, I think that’s very important to plan out that they can scale as long as your assumptions don’t change and your goals don’t change but sometimes, with the different sizes, you are focused on different goals.
Natalie: Great, so I think we have time for just a few more questions. I have one here from the audience. It says, “How many years forward can you see with prognostics?”
Steven: We have been able to show multiyear prediction capabilities and we’re giving a pretty good idea of what failures look like out past five to ten years, in some cases. So, from a total number of failures standpoint, we can see very far in the future, but there are, on individual machine basis, you’re getting a little bit closer into the five year mark on that individual failure, what that looks like.
Natalie: Great, Steven. I have one more question from the audience, “What information is needed to make these predictions? I have a lot of data but I’m not sure what is the right data to give you.”
Steven: We have quite a bit of information that we can take in. We have what we call the wish list. We can build material information, we can take into the variation of different suppliers, we can take in information on [inaudible 00:39:43] operational data, if there’s subsequent [inaudible 00:349:46] tower information, if it’s a wind farm. So, there’s operational data, there’s environmental data.
So, that big wish list is we can take into the model quite a bit of information. But what we are really focused then is how we can get the most out of that data. In a lot of cases, we are not getting all of that, which is fine and what you really have to have is you really got to know what’s in the machine, in general. So, if it’s a gearbox, who is the gearbox supplier, and what version. We definitely want to know that individual turbine operation data, so it’s the [inaudible 00:40:20] data. It’s also important for us to understand how that’s being operated in general. Then, really focus on the past, how have you been operating and maintaining that machine. So, there’s a lot of maintenance information that’s available if you did an uptower replacement and if you did some other things that are important to understand the current state of that machine. So, obviously that’s a very good input for us as well. That sums up, in the general buckets, the important pieces. There maybe a little bit more, a little bit less in different areas. If you have more specific questions on that, we can follow up.
Natalie: Great, thank you, Steven. So, I think we are going to wrap this up right now for the questions period. I have some other questions here, but I’ll again reach out to everybody and put you in touch with the rest of our team. I want to thank you all for joining us and again invite you to our next webinar, which is on July 15th. We will be hosting it from 1:00 to 2:00 again. It’s on multiphysics prognostic modeling. So, that’s a little bit more of a deeper step into our technology and how we make the actual models.
I’d also like to invite you guys, we are hosting an event with the Industrial Internet Consortium. I see a lot of companies and names that are on our attendees list today that are part of the Industrial Internet. So, I wanted to reach out. We are hosting an event in Houston, Texas next Tuesday and there are a lot of people that I think would provide some value to that event and it would be great to attend.
So, we are expecting about 100 high level executives, so please feel free to reach me for some more information on that. I’ll pass you over a registration link and some further information. Also, I will be sending out a recording of this presentation and the Power Point slides as well. So, thank you all for joining us again and we look forward to hosting you again next month.
Head of the Industrial Internet of Services
Stephen Steen has been leading CMB system design and integration within the automotive and energy industry for over 10 years. Currently Stephen is tasked with bringing DigitalClone a prognostics based technology to the energy and industrial industries.