Aptiv PLC (APTV) Earnings Call Transcript & Summary

Welcome back, everybody. Next up, we have Aptiv, an industry leader in active safety, the early stages of autonomous tech, the human-machine interface, electrical architecture, leader in connectors and certainly a leader in L4 and 5 autonomous tech largely through its JV with Hyundai. There's a ton of great product at Aptiv that's driving above-market growth, but the one that appears to really be spearheading growth for some time to come is the smart vehicle architecture as EVs become more and more prevalent in discussion and development at the automakers, plus the evolutionary path to ultimately getting there. Today, we're very happy to have Lee Bauer, VP, Advanced Engineering and Smart Vehicle Architecture; Elena Rosman, VP, Investor Relations, who is always helpful and has all the answers that we need. She's super helpful. I'm sure she's super helpful to all of you folks as well. I'm going to turn it over to Elena to give some opening remarks, then Lee will speak, and then we'll get into Q&A. So thanks so much for both of you for joining us today. We appreciate it.

Thanks, John. It's a pleasure to be here. Maybe I could just kick it off. I think many of you have heard us say before that our industry is really at an inflection point. We continue to see an acceleration in electrification, driven by more stringent CO2 regulations and actually increasing consumer demand globally. Consumers are also demanding more safety and connectivity features, which are increasingly delivered through software. And this increase in software brings with it more advanced sensing, compute to power it. And all of this has put a tremendous strain on the classical approach to vehicle architecture. And that's where we find ourselves today. Solving these challenges has the potential to both meaningfully improve and save lives. And our ability to leverage Aptiv's full stack systems capabilities is helping our customers realize their future technology road maps. So here with us today to explain all of that in more detail is Lee Bauer. As John mentioned, he is Vice President of Advanced Engineering and Smart Vehicle Architecture. So with that, Lee, if you want to give us a little bit more of the highlights on SVA and then we'll open it up to Q&A.

Thanks, Elena. Great to be here, this forum. Been here several years in the past, so it's always been a good exchange. I think what's important over the last 1.5 years is the need for our customers to identify with their user experience in a way that we've never seen before. When you take an engine out of the equation, you have to set your vehicle up to be a platform that delivers something, that delights on a daily level. So the features and the content are going to be very important in the relationship with the end consumer as well. In order to do that, our OEM customers had to address one of the fundamental problems, which really is this increasingly complex architecture that underpinned everything they had done up to this point. We were fortunate to sit in the position with the brain and the nervous system, and we could see this inflection point happening. We could see it from the ADAS controller side, from the infotainment and then up from the bottom, from the wire harness. And we said to ourselves, hey, there's got to be a better way to do this. And then we could see the trajectory of where our OEMs were going, and there was going to be more content, more features. So the challenge was going to be extreme. Another event that really pushed us towards architecture was the events of last year. Money became tight at all our customers, and they had to focus their activities. And where they ended up focusing is on electrification and on architecture. And these 2 together to allow them to create a brand-new user experience. SVA very simply was about solving the problems, we call life cycle management, and the 3 parts of the life cycle. First and foremost, in development, you've heard a lot about software delays for cars, et cetera. That's because of the way they're developed today. So SVA attacks that. It's in the production plant, where the complexity of our wire harnesses, of the electronics is reaching a point where you really should go in a different direction. And then finally, we talk about the user experience. The user experience cannot end in the plant. It has to be taken out into the field and handle things like warranty management in a better way, but also deliver the customers more features and functionality over the life cycle. We call it having a relationship with their end customers. So with the technical principles around SVA, we've talked to them, just a short repeat here, abstracting the hardware from software, separating the I/O out of the compute, moving into this thing we call zones and creating these high-performance compute clusters in the vehicle allow our customers to deliver that software-defined architecture of the future. So with that, just a little lead-in, and I'd turn it over to John to start the fireside chat.

Great. Well, thanks both of you for those comments. I mean, I guess the first thing here is, I mean, if you're redoing the complete electrical architecture of vehicles, which sounds like it makes a lot of sense, you're going to need to do that in conjunction with your customers. And in some cases, some of these customers are arguably a little bit more dinosaur-like in their approach and how they move forward with stuff. Some of them are a little bit more advanced. But I'm just curious what the customer receptivity is to this. Are you pushing it to them? Are they pulling it? How are they reacting to what seems to be something that is absolutely necessary and you're clearly at the leading edge of this.

When we started, we had the credibility with our customers because we were delivering the components. So when we first started with SVA, was in my group at the time, they knew we understood the problem. So we had a, I would say, a seat at the table. And when you expand it out into the broader Aptiv, the one Aptiv approach with the brain and nervous system, we started to stack up all the issues because the issues were happening in every one of our divisions. And then so they agreed there. So that was the first layer of confidence. The second layer of confidence is we threw out a solution. So these are very intense meetings when you're talking architecture with an OEM customer. SVA, there's generally 10-plus experts in the room talking to various aspects, from connectors, to wiring, to power management, to power moding. I can go on ad infinitum. And then as you have these engagements with the customers, they get prepared on their side. And finally, you reach a consensus of what the answer should be. So what you're seeing now are really visible announcements. You can see physical architectures that look very much like SVA. And that's no mistake. So the reception was very good before last year and has accelerated into this year as we see now most of our OEM customers needing to adopt new architectures in order to reduce costs, first and foremost, but also to deliver the features that they're going to need to, to be competitive.

So I mean, it seems like they're embracing this, but it's a tough thing to kind of do immediately in just one kind of flip of the switch. So I mean what is the evolutionary process to getting them there? Or is it flipping a switch, and hey, listen, hey, this next vehicle, whether it be an EV or an ICE vehicle, it's just going to have SVA as opposed to, I mean, and this is not to be disparaging, sort of more spaghetti-like electric architecture, which is still very valuable and very good, too, right? I mean, that's not bad stuff, but getting too complicated. How does this actually work? I mean I know you guys have talked about satellite architecture, which seems like to be a step there. There's multi-domain controllers. There's a lot of things that seem like you can partially get there. But how do we think about the evolutionary development here or is it just a revolution, I think, the switch flips?

Yes. I think you mentioned it. The first step they had to take was this domain centralization. And that was a bold step. A multi-domain controller was really the first indication that they realized that, that was the right step to take. So now you say to yourself, okay, we can do it on a domain level, there's 50 to 100-plus controllers in the car. So if I want to bring on a brand-new architecture, first off, I need to have an opportunity to do it. So battery electric vehicles have created such an opportunity. And second is, where do you start? And where you start is actually the zone controller because the zone controller sets you up to solve a couple of different problems that are diverse but related. So the first thing is, you want to fix the wire harnesses. The wire harnesses have become so big, you can't package them anymore. And going to a battery electric vehicle, you want to reduce mass. So the first feature that we brought into our zone controller, we call it Power Data Center, was smart fusing, which is digital power management. And with digital power management, we can control it so tightly that we can reduce the cross section of the copper on a normal harness today, so capture that savings in weight. So on a vehicle, it's 8-plus kilograms of savings apples-to-apples just by going to this technology. So that's the first thing you get. The second benefit you get by doing that is you can break up the harnesses into much more manageable zones. So the future of electronic architecture has nothing to do with these big body harnesses that you see today. They're going to have to be smaller, much more nimble and able to be not just manufactured in an automated fashion but also installed in the same way. And then the final piece is, that's good, now I've added content, how do you pay for that content? Well you pay for that content by turning it into then an up-integration platform. By taking some of the functionality that is distributed in the zone, could be rear lift gate, could be a headliner, could be some seat functionality, dampers, there's lighting, whatever, and you centralize that inside the zone. So then you have the economics on your side as well. So the first place you start is zone control. And I think you can see in the market today, there's a lot of discussion around zone control. We're bringing the first type zone controller into the market in 2022. And we'll see that then progress by OEM throughout the decade as the first step.

So we're getting there. So I mean, so when do you think SVA in its purest form actually manifests itself in a product? I mean, I know there's this continuum to getting there. Is it on a brand-new? I mean, and I'm not asking you to disclose anything specifically at the moment. But like is it on a brand-new EV platform where there's a pure white sheet and you can actually go in and implement it? Or is it going to be on an EV with multi-domain controllers, satellite architecture? And then over time, it finally trips into the SVA in its pure form?

That depends on OEM. What I can say is the first implementations will actually be in ICE engines. Some of the OEMs wanted to get a flavor for how to manage zone control. Not the most economic approach, but a risk averse approach. Then there is the other class of OEM that is going to bring it to battery electric vehicles first. SVA was made for battery electric vehicles. It's just a natural. When you look at the wiring and what we're doing with some of the manufacturing processes, the new connectors, certainly the domain centralization. So we'll see that then probably a step-wise migration if it's an existing battery electric. And then some of the new ones will be the ones that get most likely full SVA right out of the chute. Now what might be a little laggard at that time are the fully automated harnesses. We believe those will all happen in full by 2030.

Got it. And when you think about the land grab here of content and then the incremental content. So I mean if you're pulling away controllers with multi-domain controllers and pulling them away from functionality of other specific parts and specific areas, does that increase your potential content? I mean, you're replacing your existing content to some degree with new stuff, but you're also maybe pulling from other parts of the car. I mean, is there actually a content opportunity here for you to almost take share in electrical architecture of the vehicle or largely repricing things that you have? And then are you actually upselling at higher price points because you're providing a much better solution?

So what SVA is fundamentally is an up-integration platform. So at the bottom, we've talked about the zone controller. So the zone controller is brand-new, doesn't exist today. So any functionality that it up-integrates is, it was resident somewhere else. So it was somebody else's value. So the value we take is certainly the zone controller and then bringing that sizing for that functionality that you're up-integrating. Now it's leaving behind housing, some memory, some other controllers, et cetera. So we're sitting at the up-integration point at 3 different layers. So that's the mechatronic rim layer that has up-integration. And depending on the OEM, some of the functionality they do themselves. And they have actually inside of these smaller controllers. So in one instance with one customer, we're up-integrating 9 particular ECUs. And 3 of them, the logic inside is owned by them. So they're really excited about it because they're going to save money. So we get value, they save money. So there's going to be a lot of these types of situations where they actually participate in the writing of the code and they're going to be able to centralize it. And as you move up the chain, it gets more interesting. So then you have the central vehicle controller, which for us is the way you connect to the outside world and what brings the 2 worlds together, the high-performance compute and then what you see around the rim of the vehicle and the zone controllers. And then that's where the world will be able to expand functionality over time. Functionally, we don't even know yet because there will be a cloud connectivity, certain functions and features that can cross that barrier. So that is a up-integration platform. And then finally, it's the high performance compute. So things like advanced safety. Advanced safety is going to evolve. It's going to have to take advantage of SOTA/FOTA. It's going to bring new features along with it, user experience, interior sensing. So we're sitting really at the forefront of all of those. Interior sensing is going to advance. The cabin monitoring, it's a number of things. You're going to see a whole new layer of features happen on all 3 layers of the architecture.

And Lee, maybe I can just add to that. So a lot of the additional up-integration content is, that Lee referenced, is in our Advanced Safety and User Experience segment, starting with the zone control and extending to the CVC and eventually serverization of compute. But I think as part of John's question, right, there's also some migration on the Signal and Power Solutions side, moving from copper where we don't make money. Today, that's pretty much a direct pass-through to more sophisticated cable management schemes, engineered components that are of higher content. Lee, you want to talk a little bit about that?

Yes. So we showed, so one of the other driving forces behind SVA was to solve the wire harness complexity issue. We're delivering the largest harnesses we've ever delivered. Some take well over 10 people to install. And the only trend is they get bigger. With a traditional architecture, you add more boxes, you're going to add more of it. It wasn't going to work anymore. At the same time, labor arbitrage is coming to an end around the world. And a lot of our OEM customers are seeing, there is a point in time where if we are not fully automated, labor is going to surpass it and not stop. So they put a stake in the ground when they need to have automated harnesses, not only in our production, but also automated installation in theirs. So that was a major driver. In order to do that, and if you look at what we do at HellermannTyton, cable management has become a big thing. Today, it's clips and tape and some straps, but it's going to turn into a more advanced multi-planar type solution. So we showed that at CES 2 years ago. And then to bring it together, you're going to have to have a series of different dock and lock concepts. So we haven't sat on our hands the last couple of years. We've really built out the dock and lock. That was really a requirement because as you centralize a lot of this I/O, you need different style connectors, very dense. And in order to get them to automate, you need different carrier concepts, double-sided concepts, et cetera, even as you push it out to the rim. So there's been a lot of innovation going on, on that side in order that SVA can take advantage of it.

So I mean there's a lot of benefits, but one of them seems like, you kind of alluded to it before is OTA updates and connectivity. I mean, does this facilitate in an organization of architecting in a more, it's complex, but simplified way, allow or facilitate OTA updates in a way that might not be allowed right now and simplifies the security because you're creating one. You can kind of create one entry point because you've got more control and handle on what's going on in the electrical architecture. I mean, how does this impact OTA updates?

Yes. So today, generally, OTA has been isolated to certain domains. Our products are OTA capable in the ADAS domain, certainly in the infotainment domain. What you're going to see happen is the whole vehicle has the capability. And everybody thinks of OTA as changing out software or improving firmware. There's also another component around monitoring that's going to be very big. So as you create a fully connected system, it's understanding that it's performing well. And that opens up a whole new area of analytics. And you're going to use OTA, over-the-air updates to send some services into the vehicle to take advantage of that ability. And of course, do the basic software update and the rest that you talk about. So when you have all this together, you will fundamentally have a better car over the life cycle. And for the OEM customer, they're going to be able to have a direct relationship with the customer. In many cases, they don't have today because they go through a dealership, et cetera, because they're going to be constantly bringing new things to them, improving the user experience, making their whole thing more sticky. And the best experience of all is advanced safety. Advanced safety, you have it once, if anybody has ever used AEB or some of the other features, you want to have them again. So there's potential for them to do upsell, et cetera. So OTA, key enabler, having the right architecture in place is foundational.

So also, one of the key things is abstracting the software from the hardware, right? So I mean, this kind of gets into the OTA discussion as well. I mean, at what point along this transition does the software really get abstracted from the hardware in a way that is actually usable, right, or updatable, right? Because I mean, that kind of is like where, I mean, you can certainly save costs, simplify things to some degree if you're abstracting the software from the hardware, helping development going forward. But I mean, also one of the key things is how updatable the hardware will be with new software. So when does that actually occur in a way that is impactful?

So from a time, I'll answer it. If I understand your question correctly, there's 2 parts to it. One is when you're going to see it and then from a timing perspective, when, so it's really impactful. Before I answer that part, part of the why you abstract the software from hardware, so you have reusable software blocks. So again, I talked about the life cycle, the 3 areas that you're going after costs. So this solves a structural issue inside the OEMs. And I would say a direct benefit of doing that is the ability during the life cycle to change out the hardware. And there are several cases whereby you just love to get a new infotainment system. If I could just throw in a new piece of hardware and if I was abstracted, I could put the software on top of it and then if I had the underlying platform. So I believe that will become reality with the next-generation vehicles. The capability will be there. And it's up for the OEMs to decide if they want to do and if they want to do an after sale. But the fundamental capability from the software architecture will be there, but also from the physical architecture with things like dock and lock.

Okay. But I mean, is that the kind of stuff that's capable, that you have on your electrical architecture right now or is SVA going to need to be fully implicated or as we're going to multi-domain controllers and satellite architecture, can this stuff start happening?

Yes. So we can do today OTA, but focusing in on our like ADAS controllers, for example, in our user experience controllers, interior sensing, we can do it today. But that's really just scratching the surface. You're going to see new features and functions come. You're going to see new analytics-based functionality. That's all based on the concept of over-the-air updates.

Okay. A lot of this is very sophisticated, right? And a lot of your competitors are maybe not as sophisticated as Aptiv, for sure. So I mean, as you're looking at your wiring harness folks, Yazaki, Sumitomo, Lear, Leoni, do they have competitive responses to SVA? And who are the other major suppliers that you're coming up against as you're competing for this? And I don't know if you want to comment on the competition, but how far along do you think the competition is in this as well? And the automaker is going to look for a second source for what seems to be some pretty critical technology.

Yes. One OEM customer said to us, and if you like a tip, we were in a pursuit, he says, use your strength and your strength is the sum of all our parts. So when we go into the customer, we're not selling them a box. We're selling them a system solution. So we lay it all on the table how the whole thing comes together. So the move towards system-level solutions has always been important, but it's more important now than ever. So we've seen some competitors try to partner up with a wire harness provider. A standard wire harness provider wants to make more wires. So the motivation is probably not as high there as it is for us, if you look at from a total value perspective. The same thing could be said for the other side. So you have somebody who's delivering these boxes today and is going to lose blocks, probably not as motivated as the OEM is or certainly as we are. So we see the traditional players in the wire harness space, the ones you know, you have the Yazakis, the Leonis, the Dräxlmaiers, the Lears. More on the performance compute side, you have the Contis and the Boschs. From a full architecture perspective, no one's been able to put it together like us and for a good reason.

So if you think about this, I mean, you've got a customer base that is reasonably, I mean, seems to be fairly receptive. They're going to understand they're going to need this over time. But there may be an opportunity with new EV OEMs, right, that seem to be getting on this sooner and faster because they're creating new vehicles. Have you? I mean, what is the relationship with some of these new start-ups? Are they embracing this a lot faster than the traditional players? And what kind of opportunity do they provide you for this, for SVA?

Yes. We have relationships with many, if not all, of them today. So I would say SVA is a continuum. If somebody was putting in an architecture today, it's rolling off the line, it's probably not SVA. But I think how you could characterize it is, as we go into the next decade, the trend seems to be clear.

Okay. So they're on this and they get it. They're not creating the same mistakes all over again. I mean, they're white sheeting this and have got a good idea of how to do this probably with your help.

We had a discussion a little bit earlier today. We talked about the concept of legacy and legacy being bad. The only good thing about legacy is, if you employ it properly, you know what not to do. That doesn't mean you can change it, but you have a book of lessons learned, you know what not to do. Having a start-up, you have certainly a white sheet of paper, but you may not have those lessons learned. So it's how well the 2 of them are brought together on the inception of the first vehicle. And that will vary by company, not single out any particular company here. But as they get mature and they start to realize the power of having architectures of this type, we envision that they'll all move down this path. And certainly, they'll see some of the problems with wire harnesses, some of the problems with connectors. It's just unavoidable if you take a component-based approach, regardless of how good you are.

Yes. Got it. So I mean, as you're developing the technology, are there partners that you're working with either on the software side or maybe even on the hardware side? I mean, I would throw semis in there as kind of in between, right? I mean, it's kind of both that are critical. I mean, I know you have a relationship with Valens, and that seems like it's a very fruitful one for both sides. But I mean, is there help from the outside? Are you helping other folks as well? I mean, how is this working on the product deliverable to the customer?

Yes. No, I think Valens is a good example. As we started to look at how to connect the sensors, we realized there was a gap. And we need a different technology. And what we were looking for was how to be able to understand the performance of the link. So Valens made a lot of sense because they had a solution that I can always test the performance and know that it was available. So I could basically confirm that my sensor was, if my sensor was working, I knew that the link was working. That was called HDBaseT. While we since work with them to bring it into automotive. So they've taken that technology in HDBaseT and they pushed it into the MIPI consortium, MIPI Alliance. And now it's turned into MIPI A-PHY. We're going to see MIPI A-PHY pop up in cameras. Sony is going to adopt it, et cetera. It's probably one of the de facto standards between a sensor and our zone controller. So a lot of our activities are borne out of solving technical issues. So again, the Valens was a really good example. There are others, but we're not at liberty to say. So what I would say is, as you go into a new space like we are, we said a couple of years back, we allow parallel industries to inform our decisions, not the least of which is IoT. I won't say more on it. But the IoT space, there's a lot to learn about deploying connected vehicles en masse. And so we're leveraging some of the industries that might be slightly ahead of automotive and then some other technologies that will enable some of this high-data connectivity inside the vehicle that are required. So yes, to answer your question, absolutely.

So if we think about, I mean, and not to harp on Valens or I mean, the opportunity there. But I mean, my understanding is that they allow for a single wire, whatever type, and I'd love to hear your thoughts on this and how it works because you're obviously 10x smarter than I am on this stuff. I mean, is that the data and power can be put through the same cable or wire or coax cable. I mean, how important is that to stuff that you're working on in SVA? Or maybe even is just how important is that to simplifying the wiring harness that you have? You probably have 2 different cables that it can now replace 1, 2 or 3 or maybe many with that kind of ability. I mean, am I understanding that correctly? And how important is that over time?

Yes. So ultimately, SVA is about saving money in different aspects, complexity right down to physical cost. This one is a good example. So what we were trying to do is go with unshielded twisted pair. So shielding is what costs the money. So with HDBaseT and now what will be MIPI A-PHY, I can do 4- to 8-gigabit links, which are high-speed data links with unshielded twisted pair, and then we've developed a proprietary process on how to twist it. So those links save money for every OEM. So that was a key driver for us for the technology. Anywhere you could avoid shielding, you save a bunch of money.

Okay. One more question before we hit the last one here. We've heard a lot about wireless battery management systems which is interesting. It's a little obviously outside the realm of SVA, although it could theoretically be argued that when you start talking about wireless connections, it could compete or be complementary depending on the way you want to look at it. How does that potential for wireless connectivity in the vehicle mesh, complement or compete with SVA? And it seems like something as critical as BMS, and maybe this is my own stupidity, that kind of connectivity wirelessly, it seems a little bit risky on something that's that mission-critical in a vehicle. Just I don't know if you can comment on that and what that means for SVA or what this means in general?

From a technology perspective, I see everything in time. For what time am I working it? In the vehicle, many of things that we work with are in real-time and then are at a frequency of, let's say, 2 kilohertz. So you're getting samples at a rate that's into milliseconds to microseconds. So when you need that level of precision, it's very difficult to use wireless. Now as you move up, depending on what the precision is that you need, I can see it possible to do a BMS certainly with using wireless technology. Now we've deployed different wireless technology in the car already. Bluetooth is a wireless. How we do our keyless entry is wireless. We have WiFi in the car. So all, a lot of this stuff is use case driven. So inside the battery, there's a lot of stuff. So there's certainly an interest in trying to remove some of that stuff and go wireless. So I think there's a lot of potential. That holds some potential. Absolutely. It doesn't impact SVA. SVA is working on a different side of it. If you look at where we sit, we show up right with the battery. The high voltage, boom, we're right there. And then we live also in the low-voltage side.

Got it. And then just lastly, and Elena, this might be for you on the near, on sort of the near term is the semis issue or shortage, certainly something that's created a lot of heartburn in the industry. You guys are obviously more aware than anybody else. I mean everybody is, right, in the industry. I mean, I don't know if you can comment on the short term on what kind of an impact maybe or ways that you're working through this and the industry is working through it. And then Lee, I mean, when you think about multi-domain controllers and simplifying stuff, do you potentially use less semis and actually maybe have less draw on the semi industry and might be able to be much more efficient with this and create longer-term solutions on capacity there? Maybe, Elena, maybe somewhat your turn and then Lee would.

Yes. I'll start with that question. And then maybe, Lee, you want to follow up with the second part of John's question. So absolutely, the global automotive semiconductor shortage has led a number of OEMs to significantly reduce first half vehicle production expectations. We're working with our suppliers, with our customers to continue to monitor the situation. And really, we're seeing availability of semiconductors begin to get replenished in the Q2 time frame. And that's what informed our view back in February, early February when we gave our outlook for 2021. We knew that production upside in the back half of the year would be somewhat capped due to the shift in production and sort of production schedule availability. So at the time when we gave our guidance for 2021, we felt like the industry could recover to about 84 million units. That's global vehicle production for the year, so up about 10% on a year-over-year basis. And again, John, we don't give mid-quarter updates, and we certainly don't have Q1 guidance out there. So I don't have any near-term comments to add at this time. Maybe, Lee, if you want to talk a little bit about the change in architecture and how that impacts the consumption of semiconductors.

Yes. I think what you're going to see is SVA fundamentally has fewer, more powerful semiconductors. So that will help necessarily. The semiconductor use in a vehicle is varied and diverse. So some of the ones that are affected are probably not the ones that you would think. So with SVA, we're going to put many fewer semiconductors. So that's going to help any situation where there's...

But that's a longer-term trajectory, Lee.

It's not tomorrow, to be fair.

Yes. Yes. It can't be swapped out. Yes. It obviously can't be swapped out tomorrow. I mean, it would be great if it could for your business and for the industry. But yes, it can't. Okay. Well, thank you both of you for joining us. I mean, Lee, I always appreciate talking to you. My head spins in a very good way after I talk to you. I learn a lot on trying to figure out what I just heard and what I just processed. So thank you really so much. And Elena, thanks for your time as well. We really appreciate it. Thanks so much. And hopefully, we'll see you soon in person, hopefully.

Thank you, John. Bye-bye.

Stay safe. Thank you.

Thank you so much. Stay safe, guys. Thank you.