Brought to you by the reinvented two thousand twelve camera. It's ready, Are you didn't touch? With technology? With tech stuff from how stuff flix dot com. Hello, agaan everyone, and welcome to text stuff. My name is Chris Paulette and I'm an editor at how Stuff works dot com. Sitting across from me, as is typical of days like this, Yeah, Friday's That would be senior writer Jonathan Strickland. I looked at my notes and I didn't like them. Hey was
that a quote? What were you actually saying that? It's it's both both ay quote and it's true excellent that those are always the best, right, Yes they are. So today we're going to talk a little bit about robots, which we before we started recording, we were I was talking about robots. He was insisting on saying robots, So I didn't think you would actually do it in the episode. Of course I'm going to do it the sod. My goal is to alienate all of our listeners until no
one's left. Uh, and I'm doing quite well, I think. Uh No, so robots if you prefer. We we wanted to talk about this mainly because there was kind of a cool story and and really cool video that came out recently recently as the recording of this podcast. This is a March twenty three if you're if you guys at home or wondering, because you know, of course we published these well after we've recorded them. But the story was about a company called Kiva. Yes, Kiva Systems technically,
and Kiva Systems makes robots. They make a warehouse robot system that's all about finding and retrieving inventory using potentially more than the thousand robots if you were to have a really complex system. Now, most of the systems they've they've incorporated have been smaller than that, but the idea is really cool. And the reason they're in the news right now is because, I mean, the company has been around since two thousand three, in one form or another.
It started off as a with a different name, but anyway, it's been around since two thousand three. The reason why it's kind of hit the news recently is because another company has acquired Keyva Systems. Yes, that would be Amazon dot Com. Yes, for the princely sum of seven dred and seventy five million dollars, so just under a billion um.
So yeah, this is a big news and you would think, oh, well, that makes perfect sense that Amazon would go after something like this, like they would at least employ a system like this, if not actually go out and acquire it, because I mean Amazon is all about warehouse management in order to fulfill customer needs. So yeah, So the company actually was started in two thousand three, as we said a moment ago, by a guy named Mick mounts Um,
and he knows a lot about fulfillment. He actually kind of learned the hard way by being part of a company known as Webvan. So web Ban, for those who don't remember, was an online service where you could actually order groceries online and this company would go out and fill the orders and deliver them to your door, so that way you didn't have to go out to the grocery store. And this this was one of those companies that was was expanding very quickly just before the dot
com bubble burst and there was the big crash. So if you've heard O our podcast about the dot com crash, you probably heard us talk about web Ban. Part of the problem they had was that the cost for their system was greater than they had originally um projected, and so they were not able to bring in enough revenue to cover their costs and as a result that the company itself collapsed. Yeah. Mr Mount seems to be one of those people that learns from experience. Well, that seems
like a reasonably good thing to do. Yeah, that he I think it kind of bugged him based on what on the information that I've picked up up on on doing the research and and he started thinking, you know what, there's got to be a better way to handle UH storage and fulfillment. So yeah, he has a very technical background.
Um you know, and uh, I believe as a matter of fact, he's an m I T graduate, M I T and Harvard Business School graduates, so he had he's probably reasonably well equipped to handle this kind of technical and the business background. He also partnered with two other people, uh, Raphaelio Dondrea. Thanks there, Hoffey. It's hard for me to say. Who was a engineering professor at Cornell and then at
e t H which is the Swiss Federal Institute of Technology. Yeah, he had led Cornell to several UH tournament victories in a in robotics competition. It's like the robo soccer games. Yeah. Yeah, he he had actually helped design robots that were able to uh to to accomplish goals in a very efficient way. And so when Mounts was looking for potential folks too to work with, Dendrea became one of the natural people that he should see because of his his extensive experience
with the robotics. And then the third founder is Peter Wherman, who is a multi agent systems expert and a former professor of computer science with North Carolina State University. So, uh, and I think they were friends first, they didn't. It
wasn't just one of those like personality. Yeah. So they these these three men were combining their knowledge to try and build out a system of robots that could fill this need of being able to seek out inventory within a warehouse, retrieve it and bring it back to someone for packaging. So when you're filling an order for a customer, you're working in a warehouse and you the order comes in. A customer has asked for a particular kind of DDD player, and that one DVD player is located on one shelf
in this massive warehouse. It takes a lot of time for a human to track down which shelf it is, walk to that section of the warehouse, retrieve it from their walk back to the shipping department and ship that out. So the idea here is that you would have a robot system where as soon as an order comes in and uh and a person is ready to ship it, that person pushes a button, the robot zooms off exactly to the point where the the item is lifts up
that entire shelf. Because the robots, if you were to look at these things, they look kind of like room buzz that are on steroids. You know. They're these sort of sort of squarish robots that have rounded it rounded corners um and they are they're low to the ground and they run on wheels, and so when you look at them, they don't you know, they're not humanoid or anything like that. They what they do is they scoot around on the floor underneath the shelves and then what
they have a little screw drive that comes up. It lifts a platform up that lifts the shelves off the floor. And in fact, the robot rotates in an opposite direction of the screw at the same rate, so that the platform remains from the shelf's perspective stationary, and just it increases in height, but it does not rotate. Yeah. Now if you if you think about it, this is this is important because you have a vertical shelf. These are
essentially UM a rectangular prism going straight up. So the base of the shelf is square and and is roughly just larger than this robot. Um. But it goes up and it's got stuff on it. Well, you know, if you have things on the higher shelves but not necessarily on the lower shelves, that's going to change the center of gravity considerably. And if um, this this screw drive didn't have that counter rotation, uh, it could cause the shelf to topple or be unsteady. And the thing is UM.
When I first started getting interested in in the Kiva story and was watching the videos, I was and that's so cool watching the little bot outside spin like that. It's really cute. UM. And then I realized that it's being done for a very specific reason. And not only is it cool looking, it's functional. Functional. Yeah. So it lifts the shelf directly off the floor, and then it will maneuver back to the person who has pushed the
button to fill this order. It then shines a laser pointer essentially onto the particular item that needs to be scanned and shipped, and so then the worker would pick up that item, scan it, put it in a box, and tape it up, and then it's ready to be shipped out to the customer. UH. And then the robot will take the shelf back to where it needs to be. And meanwhile there's a computer system that's keeping track of
all of these movements. That's it's not just the movements of the robots, but where the shelves are currently located, what orders have gone out, and sort of an inventory control system. So you've got you've got multiple layers of software here. You've got a certain layer of software that's in the robot, you have a certain layer of software that's in the workers station where they are filling out orders,
and then you have the overall system. You have layers in there as well, and there's a lot of overlaps so that robots don't need to know necessarily everything every other robots doing, but the system does. And the reason for that is if you've got hundreds of robots all going through this warehouse filling out orders, then there's a lot of potential for collisions, which would be a bad thing.
You don't want like, oh, there's this one robot carrying a shelf full of snow globes, and as other robot carrying a shelf full of DVDs, and as other robot carrying a shelf full of paint guns, and the three of them converge, and then you have a combo that just doesn't work even in a Reese's Peanut Buttercup kind of universe. So it's sort of like a robot in
a china shop. Yeah, it really is. So the system has to keep track of where all the robots are, and the robots have various ways of the hecting what other robots are nearby, so that way they can all move through and reach their goals without colliding with one another.
And it's actually, if you watch the videos, it's it's captivating to see how these robots will move within fifteen centimeters of one another, get so close to each other, and yet they'll stop in enough time allow one to go through and the next one will pass on, and and you multiply that by a hundred, and it's just phenomenal to see this kind of it's almost like it's
almost like a BF ballet. Well, um, One of the things too, that the computer system is doing, um to handle this is it's not just inventory control, it's traffic control as well. But the bots also have a part in this. Now. Um, if you haven't seen this yet and you're going to go check out a video, you might have this picture in your head of a lot of robotic carried shelves running around willy nilly. Well, that's
not exactly the way it works. If you look at the videos closely, or even if you don't, you could see their patterns on the floor and you might say, you know, well, you know, they kind of look like tracks, but they're not tracks. Um, not in the sense of a railroad sort of system or tracks that are embedded in the floor. The robots must follow, right, which is which is kind of interesting too, because you know, something like that would dramatically increase the cost of installing a
system like this. Instead, Um, it's an optical system. It basically has barcodes, if you will, in the different pathways that are are you know, painted onto the floor or however they are attached. I'm not sure exactly how they
are are imprinted on the floor. Um. Anyway, the the robots have scanning system built into the bottoms of them, so they know, uh, the the UM control system can tell them what path to follow, you know, get on track number two because there's another robot coming back this way on track number one, and it tells it how
far to go over and exactly where to go. UM. And it's also very cool to watch the human employees engage with this system because UM unlike past inventory systems where a worker would have to walk down rows of shelves to retrieve a particular item or two if something comes back to put the item back where it belongs, the shelves come to them. So they are standing at
their workstation UM ready to package things up. And uh, it's it's very actually, I don't know if you've watched any of these videos, these particular videos, not just the traffic stuff. UM. It's actually kind of creepy because the the employee will take off a box and we'll scan it with a little handheld laser scanner and put it down on their station and the shelf will just drive off and the next one will move over to the shelf space at their station and wait for them to
scan the item that that belongs. UH. In the in the package. So you know, the person will put in the package and tape it up and put the label on it and send it down that the conveyor to shipping um and then you know, this next box comes out, they'll scan it and the shelf drives off on its own um. And the same thing too for returns. The shelf drives out, they'll put they'll scan an item, put
it on the shelf, and the shelf drives off. It's really kind of weird, but the employees that I saw interviewed said it's much less taxing and they can get far more work done because they're retrieving more items or or putting more items back on the shelf, and they don't have to do nearly as much work to achieve that because the shelves themselves are doing the work. Yeah, and currently the robots UM the regular production model can carry up to four which is a thousand pounds. That's
a lot of weight. It's a lot of weight, but it's still they're working. They have other prototype models that can carry much more than that because depending on what you're shipping around maybe like for example, a warehouse that has lots of masonry in it, it's going to be really lots of heavy palettes and shelves and stuff. You would have to have a robot capable caring much more than a thousand pounds in that case. These these robots
do increase efficiency. I saw some projections that suggested that a factory that uses conveyor belts and and you know, more conventional systems to get materials to the people who need to ship them out, that it might take for a typical warehouse something like seventy five workers to be able to um to to carry out the orders a certain number of orders in a certain amount of time, but in using this system would be more like twenty five.
So you've just reduced the need of your workforce significantly and by fifty people, which means savings for the company. This also brings us to another discussion that always comes up when we talk about industrialization and autumns, Asian and robots in particular, which is the concern that these systems will replace human beings and reduce the need for human workers, thus in a way contributing to the unemployment rate, which
is which is a legitimate concern. I mean there is it is there is a legitimate concern that there are certain systems involved that will reduce the need for people to work and therefore more people will be out of work. Now, this is a story that has been with us since
the dawn of the industrialization age. You've heard Chris talk about the sabo that wooden shoes being thrown into a loom in order to sabotage the loom, because these these looms were somewhat you know, not automated in the sense that we think about today, but they were following a program, and we're reducing the needs of a skilled weaver so that you know, you could have a semi unskilled worker run this machine and they could produced the same sort
of quality of work that a skilled weaver could. That raised concerns back in pre industrial Europe. Well that we've seen that whole story unfold throughout the eras of of industry and um there's been some people who say, sure, it's going to take over those those manufacturing jobs, but then they're gonna be more jobs opened up because you're gonna have people who need to make the robots and repair the robots, which is true, but it's not a
one to one. First of all, the skill set you need to work on a factory floor, and the skill set you need to maintain or build a robot are dramatically different. Uh. It's not to say that someone can't master both sets of skills, but it's just it's not
like it's automatically transferable. And secondly, you don't need as many people building out or maintaining robots as you would to maintain the number of people you need to maintain or build the robots for a warehouse of robots, there's fewer than the number of workers you would have if there are no robots at all, right, for that same warehouse. So no matter what, you still have people who are not going to have work, at least not in that warehouse.
So what this really brings to light is that there needs to be a focus in the education phase to make sure that whatever education you are getting is going to be uh something that is usable in the workforce, because the skill sets that the workforce is looking for may be dramatically different than what are being offered within
the education system. So the burden of responsibility falls on not just the students, but the education system itself because we have these education systems out there that are not necessarily offering the same sort of courses that would be useful to a person once they graduate to get a job. And part of that is because the education system is
not terribly nimble. It's a it's an institution, and institutions have a lot of of inertia gas and yet it's really hard to get them to to respond quickly to changing situations because that's just not in their nature. So we do have this issue of robots potentially taking away jobs and even jobs like not not just with the KEYVA systems. I mean, that's one element, right, but this goes across all all kinds of industries, not just warehouses. We've seen other industries also affected by this UM. So
you've got that concern. Well, hopefully the ideal UM outcome is that people will be able to start focusing on the types of skills that robots really aren't designed to handle, and that there will still be plenty of work to go around. It's just that there has to be a new focus on the type of work that you can pursue to. It really starts to hit people who may
not have a lot of formal education. Uh, they are at the greatest disadvantage in that their jobs, depending on what industry there and their jobs maybe one of the jobs that are easily converted to robotics. Uh. And they are not. They don't have a background in education where they can easily switch gears. Uh. So we're probably gonna see continued bouts of of uncertainty and and and doubt. Might as well throw fear in there to get the
whole fund in there. And uh, there's probably gonna be some some rocky, some rocky travels between now and when we reach our idyllic civilization where the robots cater to our every need and we don't even have money anymore. Okay. Well, also, you know, if you if you remember in the the on the documentary Rocky three, Uh, they do have a robot. They buy a robot Rocky I forgot about. And the uncle gets to play with a robot. It always brings
him a frosty beverage. Yes, anyway, I mean, yeah, they're they're Kiva's um by far not the only company to be building uh equipment like this, that's true. And one of the uh, you know, I've seen several I'm fascinated by industrial machinery type programs in there. Many on our parent companies networks where they make stuff in store stuff, and I just find it fascinating to see. But one of the advantages of the Kiva system UH is that it is able to store things in a far more
dense fashion UH, therefore making the most of warehouse space. UM. And it does that on a on a horizontal plane. Because really these shelves are built UH and they can be configured in a number of it s, you know, depending on what the customer needs. UM. But they the shelves are all the same more or less for for
these robots UM. However, I have seen warehouses that are very very tall and they have rigid shelving UM and their robotic systems that can travel horizontally and then vertically up to a bin that could be stories high UM and lifts it off with a forklift. You know. They have optical readers very much the same as that Cuba systems do, where there are codes on the bins and it can scan the code and say, okay, well the beIN in a fourteen level six is the one that
I need. And it goes up and with the little forklift it you know, pulls out the bin and moves the parts down to the assembly line. UM, and I find that completely fascinating, but really that that maximizes the
use of the human workers that are there. UM. So I think there is a trade off to be made, UM, because these systems do and UH make the work a lot less taxing for for those people, and it does give them UH an opportunity to be more productive at their jobs, which I think, at least for some people,
would be more more fulfilling. Also, the robots, especially in the again in the Kiva systems, because you know, I've done primarily that research for this UH for this podcast, that the systems are more quiet than the conveyor belt systems.
Some of the employees that they interviewed for one of the news reports UM said that whereas before in the conveyor areas they used to have to shout to make themselves heard UM when they needed to, you know, talk to one another, in the robotics room, they basically can
talk in a normal voice. UM. There are fewer injuries, which also seems like it would be a boon for employees, but that, of course only applies so long as the robots don't develop sentients and decided to kill all humans and they follow the laws of robotics, we should be pretty much all right, all right, um, and there's a um, you know, better climate control to which is a cost savings.
Hopefully they will pass the savings on to the workers. Well, here, the thing is the efficiencies and the you know, as you improve efficiencies, then ideally that means you can either do one of two things. You keep prices the same and your profit margins go up, or you reduce prices and you pass the savings on to everyone else. But yeah, it all depends on how the company works. But yeah, I mean it's these systems are in general, I think
they're a good thing. In the short term, they're definitely going to be something that could potentially cause a lot of upheaval, but in the long term, I think that it's you know, it's definitely a benefit. Yeah, yeah, I think so. Um. By the way, in case you're undering, although Amazon did did make this purchase, and it's unclear whether they were going to continue to allow other competitors to buy these systems at this point because it is a pretty new deal, I would imagine they would continue
just because of be another stream of revenue. But yeah, well, you know, they're never to be smart company, but who knows, they might say this is well when which case, it gives an opportunity to a competitor to come out with their own system and um people like Toys r US
and Timberland UM. Several Amazon subsidiaries have been using these devices for for quite some time, like Quidsy, which owns a wag dot com which is a pet food uh company, and diapers dot com and soap dot com uh, plus Zappos UM and they also I think Walgreens was using it as well. Walgreens uses it crighton barrel uh, stap Laze, oh, Staples. Oh. I always wondered about that. Yes, the gap sacks fifth. You have to mind the gap, especially if you're in the UK. UM. But yeah, so they already had a
pretty good client base. UM. But really that that princely some when you think about it is of seven seventy five million that what that was the second highest Amazon acquisition UM after Zappos, which it picked up for eighty seven million dollars. But Amazon has said that they're going to uh bring their total of warehouses up to sixty nine,
opening seventeen new warehouses. So this acquisition over the long haul, is going to save the company quite a bit of money, especially if they can build out their warehouses with this technology instead of retrofitting. O ye, if they building it out with it in mind so that they maximize that efficiency and uh, then you know they'll they'll definitely benefit from that. By the way, do you know what company built the very first industrial robot? The very first industrial robot?
Was it Honda Unimation Nation in nineteen fifty six. They were developed from there was a fellow named George da Vole who created the patents for a a automated robotics system and these were using hydraulics that would pass very heavy objects from one point to another that we're twelve ft apart from each other. So kind of the great great granddaddy of the Cuba Systems robots. Yeah, it also
bring heavy things to you. Well, um you know, like I said, I love watching this this equipment and especially the uh well, a lot of the manufacturing robots um, which is a little different from these logistics machines, but the ones that used the suction cups and they'll pick up say a very heavy car part or a piece of a prefab house or something like that. They'll just lift it and move it around like it's nothing. Um, which is why when they turn on us, we have
no hope. So on that done, on that time, out of here. Not a happy note, Yeah, I'm not happy. Note. We're going to uhas, Chris, We're gonna wrap this up. So guys, if you have any discussions, you'd have something that you think we should cover, let us know. Let's know on Facebook or Twitter handle there is Tech Stuff h s W or you can always send us an email. Our email addresses tech Stuff at Discovery dot com and Chris and I will talk to you again really soon.
Be sure to check out our new video podcast, Stuff from the Future. Join how Stuff Work staff as we explore the most promising and perplexing possibilities of tomorrow. The House Stuff Works iPhone app has arrived. Download it today on iTunes, brought to you by the reinvented two thousand twelve camera. It's ready, are you