Vecna Robotics CaseFlow Transforms Warehouse Operations with Automated Case Picking

Alright. Here we go. Welcome to episode eight, robot versus wild. I am Josh Kvanko, and I'm joined by two trusty pals. One of you, everybody knows because she's joined us before, and that is Rebecca Lee. Rebecca, say hello. Hi, everyone. Excited to be talking about Kidsville today. Alright. So Rebecca is on our project man our our product management team, and she's been responsible for, for the launch that we did yesterday, which is what we're talking about today called Caseflow. And then Josh Hornstein. Josh is our chief architect and technical fellow at Vecna, and he essentially sort of designed this solution and worked closely with Rebecca and our product our product management team to get this out the door. Josh, thanks for joining us today. Thanks for having me. Awesome. Okay. So the title the title of this webinar is Unicorn in the Wild, True Case Picking Automation is Here. And, for those of you that didn't know, we just announced yesterday a press release. Check our website. Check our LinkedIn feed. Wherever it is that we communicate, we've announced that we we launched, a new solution called Caseflow, and it and we think it's a revolutionary approach to automating case picking. And we're here with you today to talk about it. And you've got the folks who literally made it happen on the line here for us with with us for half an hour to just peel it back and get into the whole who, what, when, where, and how, and why. And so let's get right to it. So as we usually do, I'm just gonna tee us up here. A little bit about automation technology maturity. There's technologies that are ready today, and largely speaking, horizontal transport cube storage. Technologies that are maybe aren't quite ready, except for one of them, case picking has now moved over to the left. We believe it's now ready for prime time. Isn't that great? It should have a graphic that sort of pulls case picking over to the left, finally ready. And then technologies that, are very popular, have a lot written about them, a lot of froth in the water, a lot of advertising, but we you know, humanoids being a good example of them, but we just don't believe they're ready for prime time. Here's the current state of automation adoption, as Gartner sees it. And I think, you know, I think the key takeaway here is eighty seven percent of the market has not fully deployed these types of intralogistics technologies in any grand scale yet. And, you know, this is this is our moment. This is our moment to come out with full blown solutions that just work, that meet end to end needs like case flow, and and really drive adoption into the market, because for a variety of reasons that I'll that I'll get into, it's really it's really a time to adopt these technologies at scale. So if you're one of those that have deployed the technology or have engaged in your first deployment or just beginning implementation, after you've done that, then there's, like, that, you know, that IBM commercial, oh, no. Now what? You know, what do I do next? How do I think about adopting these technologies in a meaningful way at scale? And that's really sort of the the fulcrum of the series, robot versus wild. Once the robots get in the wild in any meaningful way, you know, what happens and what can you expect? Okay? And that's what we're here to really talk about, vis a vis case picking. But in general, some of the pitfalls to scaling automation are really here on the left. You know, poor requirements definition, underestimating solution and flexibility, and so on and so forth. Weak goal alignment, fail failure to forecast changing warehouse landscape over time, which really talks to, like, are you picking the most flexible solution? And, and so as you're heading in and as you're thinking, hey. I've done that pilot or I've done that first engagement. How do I think about this technology in the years to come? And here are some some of those pitfalls that I think you need to keep in mind. Okay. So let's get to the topic of hand. I did the the business end of this. And so now we've seen we're we're teeing up this concept of case flow as a as a unicorn. And and and we call it that because we think it's true case picking automation. And there are other technologies out there. We believe that they're either, they hit the mark, but perhaps too inflexible and, you know, way too distracting, right, if you will, for lack of a better term. And then there are other flexible solutions out there certainly, like voice picking and others, that may be efficient, may be less expensive, but don't quite hit the mark. Right? So I have Josh and Rebecca here to peel us peel this story back a little bit and talk a little bit about more about why we think our solution is is that unicorn. And, you know, in case you're you're like, why did this come why did Vecna choose case picking as a, a solution to solve? Well, I mean, it's predominant in warehousing and distribution. And if you look in the middle the middle bar here, most of the activities involved in case picking, really could benefit from automation and really should be bent, automated. And then if you migrate over to the far right, most of you out there, not most of you, but the plurality of you, forty five percent, are planning to automate this workflow in the next eighteen months or less. So there you go. It's really a great time to be bringing solutions to market that address this critical need. So with that, hopefully, I've done a good enough job, to hand it off to Rebecca. And she'll start to, uncover why we started tackling the solution, as opposed to, you know, a plethora of other automation solutions that we could have, brought to bear. Why don't we start with case picking? So, Rebecca, I'll hand it off to you, with that question. Yeah. So I think Josh already kinda hit on a lot of, the good stats that supported, the the the bullet points here. Like, first, it was still a very human driven workflow today. And if you remember from the previous slides, fifty five percent of all the time that people spend doing is traveling. And if you think about kind of the different activities inside, that compose a case picking, What humans really good at is the picking part. Right? And so there are a lot of kind of opportunity for automation to step in and, really help reduce the inefficiency and focus our precious human labor on the most value creating task, which is picking. And and it's like, when we started talking about case picking with our customers, it was very clear to us there is a strong demand for automating this. We're bringing a more flexible automation solution to solve the problem today. And another thing we kind of unveiled was that the current process, because it's so human driven, it's so labor intensive, they're often lacks, like, a system level data capturing and system level optimization. So a lot of those kind of then become obvious to us that this is a problem worth solving, and, it's a problem that we can solve. And we already have a lot of the pieces ready with our existing technology to tackle this. And this is probably something that, Josh, you can, comment a little more on, the existing things we already have. You're on mute, Josh, by the way. Thanks for catching that. Mhmm. We have, like, today already, the autonomous platforms. We have the pallet trucks, the tuggers that are able to go from point a to point b. The technology that we needed to build for case flow takes that and just allows us to use that without adding any more complex, work that those platforms have to do and building, an adjacency to that, with software, with the ability to be able to now look at coordinating humans as well. Mhmm. Yeah. So from a product perspective, we need to validate three things. Right? Feasibility, is this something that we can solve? Viability, is this something worth solving? And desirability, does the market actually want it? And I think when we checked yes to all of those questions, then it was really obvious that we should go ahead and tackle this problem. Right. Very good. Not to mention the data I just showed you before, which is, like Exactly. This is happening everywhere, and the the majority of people, slightly less than the majority of people, are looking to automate it now. Right? They're searching solutions, and all that triangulates into a great fit for Correct. Certainly for for someone who invents these things and then somebody who, who brings those products to market. Alright. So introducing Caseflow. A little video here for you all. Rebecca, you wanna give a little soundbite underneath this? Yeah. I mean, this is not, like, our official log video, but I think this is a really fun kind of in person demonstrating how it actually works. Right? The fundamental concept is that the robot takes the pallets autonomously throughout the entire journey. So a the robot actually takes away all the travel while the person still picks to the the robot. But you will notice that as, this person, actually, Josh here, finishes picking. He can just walk away from the robot. He's not attached to that specific pallet or pick list anymore. Once he finished this task, he'll go around and find the next robot while the robot itself autonomously drive to the next pick stop. And this is one kind of example of at one pick stop. And as you kind of imagine to zoom out, there are, like, maybe twenty robots all taking palettes, and there might be five or ten people all picking to, to the robots. And all of this is coordinated orchestrated by our orchestration engine in the background. Yeah. So, that's the gist of it. Right. So, Rach Josh, go ahead. Mhmm. And then getting to build the solution and being able to go out in the real world and and, like, use it and play with it and be like, okay. Did we get it right? Does this work? You know, that was really, you know, a great part of how we we build our technology. Mhmm. Exactly. Okay. So who we've talked a little bit about case flow. We showed you a little a little quick quick demo of it. Who are the cast of characters? So let's slow down a sec and say, who are the characters in this play that make this make all of this go? Before we get into the details of the the complex perhaps the the complexity behind it all, what are the who are the cast of characters? Right. So I think probably what people saw first was the robots. Right? So the green robot is what we call a cobot pallet jack. It's designed to operate very similar to your traditional pallet jacks, but it's autonomous. It's fully autonomous, able to take the pallets, throughout the entire pick list, and the autonomous perform a drop off. And it's designed to be able to, collaborate with a human member effectively efficiently. And then we have the operators. They, are, equipped with a wearable device that directly talks to our system, and this guides them through the entire pick journey where they're meeting the robots, all the specific instructions for the picking. And so it's like, it takes a lot of mental load of the operator. So don't have to think and plan their path, but they just, you know, get instructed where to go next and what to do next. And then, if we go counterclockwise, monitoring. So this is, I think, one key differentiator for Vecna, where we have twenty four seven command center that addresses those edge cases when the robot runs into a blocked path when, something a spill. Right? Something that's happened, that's and then we can quickly step in without, the site even noticing. And then going up, we have console. Console is the one, kind of single pane of glass where you be able to monitor everything that's happening on the floor, see all the key metrics live for your, associate performance, for robot performance, all in one place, and historical analytics data. And last, what I think most importantly is actually our orchestration engine. This is the brain behind the entire operation, a orchestrating not just the robot, but the human, where the humans are going to meet the robot and operate orchestrate the entire system to, for a system level efficiency, not just for the robot, not just for the people. And there's a lot of complexity, which we'll talk about, we can talk about later. But that's the key part. Mhmm. Right? It separates this from any other technology that you can go ahead and search, you know, Google search for case picking automation is that the orchestration software and how it's managing not just the technology, but also the associate. That's the breakthrough. Right? Right, guys? Yeah. Yeah. I mean and like like Rebecca said, we have a slide coming on later. We'll talk about how that kind of works. But, yeah, that is the the secret sauce. Right. Alright. Let's keep going here. We're making good time. Alright. So in the spirit of how was it made, that's a good title for another a new series. Maybe next year, it's gonna be called how how it's made. Do you guys like that one? Yeah. I like to go. But that's for a different time. So how did we tackle the problem? So let's let's just discuss that a little bit. Yeah. We won't go into kind of the details of, like, how we actually engineer this problem, but, like, I wanted to share some, principles that as we started to think about, okay. This is a problem we want to solve. How do we go about solving it? I think the first realization is that with the current, kind of traditional case picking workflow, you have this one to one attack association between a pick list and a, a person. Right? And and what that limits the system to do is, it it kind of remove it's a linear system then. Like, it removes the possibility for optimization. So anything that you can do, for example, like, with the worst picking is marginally improving one small component of, of the process. So with the new solution, we'll have to break this one to one association. So I think that's the first thing that we realized. And and then we take, a look at with existing technology, the robots we have today, what does it do really well? It's able to do autonomous traveling really well. It can take pallets autonomously from any one's a to b location really well, but it doesn't necessarily do picking well. Right? There's so much uncertainty and complexity to just a simple action of picking. So we want to focus the human on that task, which is challenging for robots, and focus robots on the task that it does really well. That's principle, I think, that number two. And kind of tying those two together is pretty natural to recognize that the concept zone picking, even in, like, a traditional, warehouse setting, will be, very beneficial to to improve, efficiency, and there could be, kind of drastic improvement when we introduce those two piece those two, part together and introduce orchestration on top of it. So I think those are the kind of guiding principles as we started to design a solution around this problem. Very good. Very good. Okay. So what I think it's it's interesting. You know, as you peel back as engineers tackle a challenge and you you sort of peel it back a little bit, and you're like, the beauty of discovery. Right? And and I think it's really interesting when we've been preparing for this webinar, both of you have mentioned things that that were were surprising to you. Mhmm. And I think they shed light on the problem that needed to be solved, but also how we solved it. So I've asked you both to come with some, examples of what surprised you along the way, with bringing Caseflow to market. Yeah. So I think unlike our previous, kind of, standard workflow taking just having the robot taking pallet from a to b, What, case picking is different is that there are a lot of ash cases. Some of them arise from the fact that it is a very human centered, activity, and some of them arise from just, like, intrinsic edge cases. Like, when you go pick a case, there could be insufficient quantity. The inventory is low where the product is damaged, where the barcode is damaged. And there are a lot of, like, I think, nuanced process in that workflow. Like, you need to scan a skill, where you scan a serial code, where you scan it's called LPM. Right? Like, there are a lot of variability in in this particular workflow. Yeah. And the thing the the second one was something, I think in the hindsight, I was like, why why hasn't this been solved today? It's like we have each picking automation that's pretty, I think, popular now. Like, we have, a lot of different type of automation technology in the warehouse, but case picking is such a large, kind of ubiquitous, workflow, but it hasn't been solved. Like, there hasn't really been a good solution on the market. That was really surprising to me at least. And Yeah. And Mhmm. Go ahead. I was gonna say, like, the you know, one of the things to kind of those first few points was, like, well, we have to build this and, you know, we have a lot of the pieces already, but, you know, how do we make it flexible? Right? Different customers have different requirements. Like like we said, like, when it's a human centric thing, there's exceptions that have to be handled. And so a big part of kind of building it was thinking about those and making our system very flexible. But one interesting thing that that really surprised me the most was our orchestration engine's ability to solve this problem. You know, we had built the engine with the capability of, you know, trying to say, okay. I've got this big problem of of different, of multivariable complexity, but adding humans in wasn't like this really hard thing. It's just another resource that this this engine now had to manage and and coordinate. We built it to be able to deal with the concept of how do I get two things to meet at the same time at the same place. But this is the first time where we're actually getting to exercise that logic that we had. Yeah. But just how quickly and smoothly that all worked, was was a big surprise and shock for us. Good. Good. Good. And there are questions are flowing in here. Everybody wants to know so much about how it works. You know? So we'll get to that right away. So we're not gonna ignore that. And clearly on the right here, I don't think we need to belabor this point, but, like, it's tremendous inefficiencies in how this workflow operates today. And the one that I think is overwhelming in my mind is is that that is the lack of systems level thinking. Right? I think that's the one that I think we bring some systems level thinking to bear here, in the magnitude of this of the of the problem and the touch points that happen in a larger operation, I think that's where you can really sort of unlock some major improvement. Right? Okay. So let's keep going here. So this will answer the questions, and there are two. Let me repeat them. So, Josh, let's make sure we're being clear to answer these questions. One, how does the picker know where the robot is, where there is a robot waiting? And then the other one is, how does the associate know where to put the cases on the pallet, or does it go back to a central sort before wrap? Okay? Sure. So these are probably parts of the same same genre of question here. Right? Yeah. So this this is kind of peeling back a little bit of kind of what goes on in the system and and how do those things happen. So, you know, during the day, what we call pick lists or or, you know, orders come in, and those feed into our system as what we call demands but work requests. Hey. I've got these set of, picks to do and whatnot. Our system takes that and says, okay. I've got all these AMRs to be able to allocate to do that work. What's the right order to do that? And then once I've allocated that, I've now got pickers who I've got to get to those pick stations or pick stops. And, really, our orchestration engine is its goal is to optimize throughput but by optimizing flow. Right? We're not gonna spread all the work out evenly and make people walk long distances to get between the different robots. We're gonna create a a good flow of the work so that people are near where the next pick that they have to do. How they know where to go is that they have that wearable device that Rebecca pointed out earlier, that's on them that's basically telling them, alright. Here's your next task. We don't just give them, like, a one at a time task. We look ahead and we plan the system out to look at an optimal solution and and minimize their walking distance. So it'll give them a set of up to maybe six tasks that they that they need to do. We don't force them to have to go in that order. That is just the optimal order that our algorithm figured out for them. But they can go and and do any of the tasks. We also have the ability to flexibly, like, allow them to scan a robot that's not theirs if they walk by one that, that they see is available. So it's that wearable device that's really kind of communicating this to them. Hey. Here's where you should go next to your pick. They don't have to walk around and look for robots that are flashing the light that they're ready for for a pick. Right. As far as placing the the loads on the pallet, that's not something that our system tries to give them, an optimal solution towards. That's maybe a feature that that we'll look at building. But they're pretty well trained on how to build the pallets out themselves. Right. And as if a customer needs, the system to go to a central sort before wrap or if it needs to take the the load to wrap, our system is flexible in being able to work with the customer on their workflow of what steps are after this, and we can make sure that we're dropping it off the right place. Right. So let's let's be clear about this one. Right? Mhmm. Where does the picker how does the picker know where there's a robot waiting? Software. K? Software does makes the decision, and then the wearable tells them where to go. Okay? That's number one. Number two, how does the associate know where to put the cases on the pallet? We're not solving, you know, the the pallet placement part. Right? However, because there's no driving, there's no driving anymore. Right? The pickers are picking in these dynamic zones and are told where to go. Josh mentioned there's some discretion there. They have extra time to place the the, the, boxes properly and build a pallet properly. And you heard Andy Johnson if you were joined us last month on this webinar with with Rebecca. Andy Johnson from Jodas mentioned this was a a really big upshot for them, benefit for them by deploying Caseflow was that their workers had time to build a a pallet properly, not only to to pick more, but also spend more time to make sure that the pallets were assembling because they didn't have to worry about getting on and off and and the and the travel. I also wanted to add that most warehouses, they slot items according to crushability, weight, and, because we just we follow the pick list order that, we receive from the WMS, The we'll go get you those heavy item first. We'll go to those high runner items first. However, the warehouse slot them. And, so, typically, the, the operators, they will just stack keep stacking on top, and that's, assuming kind of the WMS already take into account of crashability. That's the order that it will be placed. Very good. And and you bring up a a good point, which is the WMS. Right? This problem isn't a problem that we solve completely on our own. The coordination with the WMS, and it gets to, you know, the question that was asked about active or reserve locations. So, you know, we know the locations that the WMS is asking us to go pick to. But if there's an exception, I got to a spot, there's not enough inventory there, we can report that back to the WMS and they'll be able to send us, okay. Now go to this reserve spot, and we'll just add that to the list of, big stops that we're gonna go to. So it's a coordinated effort between our system, which can coordinate the the humans and the robots and the WMS, which knows where inventory is and where where items are. Mhmm. And and this coordination between WMS and our system all happens behind the scene. Right? Without the associates or the ops manager having to step in, it's all taken being taken care of, automatically. So with that, for example, with that retry for the exception, that retry robots will show up on a stop and the associate will be directed to pick to that robot without even having to know that that was a retry. So all of those kind of take away the mental load and have, like, be needing to plan, away from the ops manager or from the associate themselves. Okay. And and and just to preempt the question that might come up, you know, not all decisions can be made automatically. And so this goes back to the console application that Rebecca talked about, which is showing, you know, not only here's the flow of how things are going, but any exceptions that are reported, if there if that can't be automatically handled, then a manager can come in and basically say, okay. Here's how I want this one to be handled. Mhmm. Okay. Good. Do we is there do you wanna sort of wrap it up here, Josh, with going through the rest of the workflow, or you think we covered it more or less? I think we covered it. I mean, the the the big thing is, like, right, you've got the coordination of the robots doing the work. You've got coordination of the humans to work, and then figuring out a solution that optimizes that flow. That's kind of the the the big goal of, like, how do we deal with all these spots that I have to get everything to and do it optimally. Yeah. And I also saw in the q and a, there are a lot of good questions flowing. We might as well just address them while we're on this slide. So I think one of them was that, what robot devices can be integrated for this type of solution? I mean, we mentioned CPGA, but, we are able to so this caseflow software system is really not there to the robot platform. So whether it's a tugger or it's AFL, it doesn't they they all work well with the software system, essentially. Yeah. But let's be clear. Our our robot. Robot. Yeah. Right. Our robot. It will need to talk to our orchestration server, and that's why it has to be our, robot. Yes. And so for today, we're launching this on our Cobot pallet jack. Mhmm. Okay. And Rebecca's the comment more about our forklift and our tugger. Mhmm. Very interesting one here. I wanna make sure we drop sorry. Go ahead, Rebecca. I I was going to mention the second question, but, if you have comments about previous one, you should go first. No. No. It's the one about dynamic. What do we mean by dynamic? Exactly. Very Mhmm. Interesting one. And it's certainly around the company. Everybody has their own little analogy and their own their their words that they use, that fit in their brain. But, Rebecca, why don't you take that one? Can we go to the slides with the, cartoon drawing or the, This one? The the robot flow. Which one? Yeah. This one. That one. So latency is true. Probably see kind of in the background there, like, the boundaries of the zones. And it's correct that the zones are dynamic. What that means is, it's not a rigid box, that a person is confined to, but rather, it flows it's very fluid depending on the workload. Right? So what our system is trying to do is to optimize the overall system efficiency. That means increased peak density, reduce, human travel, while, balancing congestion. Right? We don't wanna cause a big robot congestion. So while trying to do that, you can almost kind of imagine this, like, fluid dynamical around the person. So if the work reduces in one area, that person will be relegated dynamically to a different area. And if someone is working a lot versus another person is maybe traveling, quite long distance and not picking a lot, we might, you know, switch them up where, so so the zones around a person is very fluid. That that's kind of the best that I can, try to put, like, some, like, imaginary image for people, but, I don't know if there's anything else that Josh will run on that. Yeah. I mean, I mean, I think the the other things to point out are the the concept of being dynamic is that it's constantly looking at the information that's available. Right? If another worker gets added to the system, it's going to dynamically adjust to best balance that issue of that worker. Or if a worker has to leave, it goes on break, it's gonna do the same thing. Likewise, as work comes into the system, it's not a system that a solution that only is able to compute it at the beginning of the day. Here's all the work, and that's that's what we do. It's able to deal with the, disturbances of new work coming in, things taking longer than expected, and so on and so forth to keep on optimizing and adjusting the solution both for where the robots need to go and what zones the the humans are working. Right. But, clearly, one of the parameters here is to ensure that the workers aren't walking twenty miles a day. You know what I mean? Like, that would defeat the purpose of of, of the efficiency that we're gained and productivity that we're trying to do here. Right? So Right. That is a parameter. It's a clear parameter. But to your point, Rebecca and Josh, the the zones are more like a lava lamp. You know? Yeah. They're they're mooning around, dynamically to adjust to the needs of the, of the system. Mhmm. That's accurate. Yeah. Okay. We're running over here, but the questions are great. I mean, this is more questions than we normally get. And, mark my words. If we miss any of your questions because we run over, I will personally answer them and email you with an answer from, from our experts here. But we've gotta we gotta make, make some time here. So, Josh, why don't we just give our our analogy, because, you know, clearly, first run through, it's hard to internalize something that's new like this, all the aspects. I'd encourage you all go to our backslash case flow, our case flow website at beckon robotics dot com backslash case flow to get a a clearer understanding of it. But, Josh Josh spent some time trying to put an analogy together that might resonate with you all out there in closing. So why don't we, take a stab at that, Josh? Sure. Yeah. So, you know, to think about, like, a real world problem that, you know, I deal with on a daily basis, I've got a family, and I've got to figure out how to get my kids from school to after school activities like soccer or dance or or whatever. And between me and my wife, we're trying to balance that, and so far, luckily, we haven't ever left our kids stranded somewhere. But for this problem, think about that kind of thing and expand it to you know, we live in the Boston area here, the I four ninety five loop. Imagine a system that could basically take everybody trying to coordinate all of their kids everywhere they need to go, all the activities that they need to do. I've gotta you know, I can't leave work till five today. I've gotta go pick up the the groceries at this time or whatever. And look at looking at that and saying, okay. Here's all the actors that I have, the people that the the resources that I have that can help me solve this problem, and here's all the coordination that needs to happen. That's what our system is doing. It's basically able to plan all that work, or all that, coordination without leaving any kid stranded somewhere because there wasn't someone to pick them up. And there are exceptions. Sometimes you need a last minute carpool help, and the system can handle exceptions. Right? Yep. Yep. Exactly. Yeah. And that's, like, the thing. Right? Like, you you get that call from your kid. Hey. You know, I gotta stay here another half an hour. Okay. Oh, how do I deal with that? Well, the system's able to take that and be like, okay. I've got a new solution. Here we go. Right. Mhmm. Very good. And so bringing this analogy back to the warehouse, like, in one shift, when you think about the complexity of doing this four ninety five thing. Right? Every household in four ninety five. I don't know how how many houses there are. There's probably a million houses or, you know, at least three quarters of a million houses in this in this area, I guess. So you take that back to a warehouse, Rebecca, and over the course of a shift, like, how many how many permutations are there? Permutations and combinations are there of tasks that could happen over the course of one shift. Yeah. This was very interesting. So I did, like, a back of envelope calculation just, like, to optimize. And, I think earlier, what I said was, like, we want to break this one to one association. Right? So that introduced possibility for optimization, but how many possibilities there are? So, like, let's take, a regular, like, two million, unit annual units warehouse. So that kind of if you spread it across fifty two weeks, two shifts a day, that's about, and if it would seem like thirty units per pallet, it's about a hundred a little more than a hundred pallets per shift. Yep. And or a hundred orders per, per shift. And in a traditional warehouse, you, like, just print out a hundred order sheets and kind of hand them in order to your, associates, and then they go on picking. But in a caseflow operation, because we kind of break the atomic pieces apart and say we have ten robots in this warehouse, they're for each order, you have ten choices to assign which robots and when, of the to assign the order, and that's ten time ten to the hundredth power. Right? And, I don't know how many trillions there are. Actually actually, I looked it up. It was ten to the a hundredth is called Google. I don't know what whether that's, where the name Google comes from, but it's called a Google. So that's a Google possibilities. And then on the other side, it's not just assigning works to the robots and also assigning all the pics to your people. A hundred orders and each pallet has thirty picks, then that's three thousand pick, picks. And say you have, three warehouse workers, and that three thousand, to the power of three, that's twenty seven trillion possibilities. That's within one shift. Right? And One shift. Twenty seven trillion possibilities. Exactly. And and that's just assigning which picks goes to which person. And so what our orchestration engine is doing behind the scene is trying to find the optimal solution between that ten to the hundredth order, different possibilities to assigning tasks for robots and the twenty seven trillion possibilities to assign, a pick to people and constantly looking for optimal solution. It sounds a lot, but I think that is also opportunity in our system because that's where possible like, a possibility for a more optimal solution lies. Yeah. We like office. Yeah. I mean, we're just trying to give you guys an order of magnitude here of, you know, what humans using a linear process can handle Mhmm. And then what software can handle. When you're getting into the twenty seven trillions right? And I'm no engineer. When you're getting into the twenty seven trillions, sounds to me like that's something that software should be solving for you. Right? Yeah. And that's, that's where we are, today. Okay. We're running short on time here. I'm just gonna sort of conclude with, you know, look, Caseflow offers a a whole host of benefits here relative to other automation solutions. It it uses existing infrastructure. It automates most of the travel. It optimizes all the resources. We've talked about the monitoring. It's really easy to set up because it's using existing infrastructure. You need less pickers in these dynamic zones, and that's where the value comes in. And it can be simply scaled, up or down. And, you know, here's their ultimate value that it delivers. Right? It does, robots do the travel, and, you know, the payoff is you need less workers to do the picking. The ROI is clear because we offer robots as a service. So, you know, we're we're we haven't run into a scenario yet, with a with a prospective client where, where the ROI was less was forecast to be less than twelve months. And then, the whole thing and maybe this is a future webinar, Josh and Rebecca will re really talk about how a technology like this with twenty seven trillion possibilities in one shift really does enable you to turn your operations into a data powerhouse. Right? A treasure trove of data that you can learn, learn from and improve your operations. And so once again, hop on to backslash case flow and, have a look at our ROI calculator. We have we developed the case flow ROI calculator just for case flow. And so you can just determine whether whether our solution is a fit for you and whether it provides you with the return that you're you're looking for. If you are interested in webinars like this, we've got we record them all. We've got three years worth of webinars on there. Thank you, Clint, for all you do there. And then, please follow us on LinkedIn. We do a lot of, marketing and put a lot of good content out there, and follow us. You know? Give us give us a hand, and and maybe some of the material we post there will be of interest to you. So with that, Rebecca, thank you so much. Really appreciate your help today. Of course. Always a pleasure. And, Josh, always great to have you. You know, always great to chat with you, but great to have you finally on our platform. Thank you so much. Thanks for letting me join in. Of course. Of course. So with that, on behalf of everybody at Vekner Robotics, Josh Kamenco signing off. Until the next episode of, Robot versus Wild in November. See you later, everybody. Thank you. Thank you.