Iowa Swine Day 2023: Evaluating, Analyzing, Implementing New Technology for Swine Systems

Iowa Swine Day 2023: Evaluating, Analyzing, Implementing New Technology for Swine Systems

Show Video

- Our next speaker this afternoon is Dr. Tom Stein. Tom Stein is the Director of Animal Health and Ventures for Merck Animal Health. He is the creator of the PigCHAMP software, and co-founder of MetaFarms.

"National Hog Farmer Magazine" named him one of the top 50 men and women who truly made a difference in the U.S. pork industry over the last 100 years. In 2011, the American Association of Swine Veterinarians presented Dr. Stein with the Howard Dunne Award for outstanding contributions to swine production and health. In 2019, Dr. Stein received the George Foxcroft Award

from the Banff Pork Conference, for rigorous and high-profile research which contributes to improvements in pork production efficiency. Let's welcome our speaker. (audience applauds) - Thanks a lot, happy to see a lotta old friends in this room, and I appreciate the invite.

First time I've spoken at the Iowa Swine Day. And so first I have to explain that I was with Maximus for five years after I left MetaFarms, and Maximus was acquired by Ingersoll Rand, and so the owners left after about a year, and I decided I was gonna leave. The same week I decided that I was going to leave Maximus, this position came up with Merck Animal Health. And it's in the Animal Health Ventures group. And I thought, "This is a great opportunity," because, so my job is, I go around the world, looking for technologies, practical technologies for pork producers, and companies that are doing that, startups, companies that we might collaborate with. So the Animal Health Ventures group is the internal venture capital group for Merck Animal Health.

So there's that piece of it, and then the other role I have is putting those technologies together with the R&D team, to put something together that we can deliver for pork producers that you'll find valuable and implementable in your systems. So it's a fantastic role, and I really appreciate the opportunity to join Merck Animal Health, great company, and also to sort of cap the career I've had with this opportunity. So I've learned a lot, really.

I joined Merck Animal Health Ventures last October, and so since then, it's been, you know, 200 miles an hour, kind of ramping up on what's going on all around the world. And not only that, the group I'm in, which includes John Kolb, many of you know John from Iowa here, but it also includes all the other species, including dogs and cats. So these are people within the group that I'm in, small group, that they're looking at what's going on with dogs and cats, technology-wise, what's going on with, in poultry around the world, in aquaculture around the world, beef cattle, dairy cattle, cow/calf feedlot, and pigs. Of course, chickens and turkeys and the rest.

So it's a great group, and so we have this cross-fertilization of who's doing what, what do you see, who's doing, what technologies are emerging, and so that is kind of coloring the presentation I'm giving today. So, I just wanna say a couple things about maybe, Dan's, I love listening to Dan. We speak the same language, not French, money, and so I always enjoy listening to you.

That question about what percentage of trials are successful, you know, Ashley DeDecker from Smithfield, who's their head of production, R&D, says they do maybe 50 or 60 trials a year, and six to seven are successes. So, just to kinda give you a sense of how you have to decide you're gonna spend money, and the failure is as important as the success, right? And by the way, so this is an art installation from, I saw it at the Renwick Gallery in Washington, D.C. So that's like 12 feet tall and 15 or 16 feet wide, and I just thought it captured everything, really, that I believe about the changes and the way technology changes the swine industry. I started, I worked on about a 800-sow operation while I was going to vet school at the University of Illinois, for three years. I started at the lowest of the low.

I was power-washing and scraping manure, and hand-carrying feed to, you know, Art's Way feeders and that kinda stuff. And so, you know, Cargill modified open fronts were the state of the art back then, right? So we've, now it's a quad, you know, quad wean-to-finish barn with tunnel ventilation, et cetera, et cetera. So you know what the changes have been. So keep in mind that, and I'm, I have said that what we're doing, my generation, what we're doing is passing the baton to the younger generation. I mean, with PigCHAMP and MetaFarms, which I was pretty involved in, you know, we created benchmarking.

But benchmarking, a lot of the answers are not in the benchmarking databases, right? We're passing the baton now to the younger generation to find insights and action, in terms of trapped value. And I'll get back to that idea of trapped value in production systems. So I want to do two things in this presentation, and I've got, of course, I've got like 110 slides. Not quite that many, but I've got plenty of slides, so I'll go through pretty quickly. But I want to do two things.

One is, I did an analysis of publicly-available information, interviews that Laura Greiner did with a couple people from these different production companies, and I did a compare and contrast of the different approaches that these four production companies, publicly available information, things that they've published or made statements and interviews and all the rest. And they take different approaches to production. But I can't go into all the detail. I'm gonna make sure to get a couple key pieces of that, and then I'm gonna dive into particular technologies, and give you an idea of what's going on out there and what's happening. But when I was with Maximus, I started writing a newsletter, a weekly, not really weekly, ended up being every couple of weeks, that I called "Brain in the Barn," and so it was me like posting an article or an analysis or a newsletter on, essentially like Substack, right? So an emailed newsletter called "Brain in the Barn," and I did this analysis. It's available, all you have to do is Google "Brain in the Barn," look for an issue number 60, and you'll get my entire analysis of the different approaches that these companies are taking to R&D, right? And then one of the things I'm gonna do is go jump into some of these new technologies, but that list could be double the length at least, of what I've got up here, so I have to say, I want to talk about stuff I know, I mean, that I've had some experience with and that I'm familiar with, and leave out a bunch of stuff that I probably should talk about if I had more time, right? So, one of the fellows that I've borrowed pretty heavily from is David Rosero.

He's been the most open at publishing or presenting the results that they've gotten on the R&D side at Hanor, and you're gonna hear more of his results a little bit later, 'cause he graciously shared his slides with me. He gave two talks this year, one at AASV, and another one at Midwest Animal Science meetings that were fantastic presentations. But I just wanted to start by saying, you know, he's talked about what it is that they're looking at, and I don't know, hopefully you can read some of this. But okay, so these are some of the big things that they're looking at, from a technology standpoint, and trying them out, right? So, you know, environmental sensors, and you've seen how BarnTools has emerged as a great monitoring tool, Distinct is out there, and then you've got all the control companies, like Maximus, or Fancom, or AP with Edge, and et cetera, right? So, the big thing that's happening with sensors right now is people putting in not just one temp sensor, whatever, but maybe 10, or 15 or 20, in barns, especially poultry, not so much pigs, but especially poultry.

Feed bin sensors, I'm gonna talk a little bit more about that, not in-depth, but just, hopefully, I mean, I assume that many of you were at the Expo. How many feed bin monitoring technologies did you see there? This is exploding, sort of a big exploding area, and they're trying, they put, I know that Hanor, they were, in fact, all the first feed bin sensors that people have started using, in the poultry industry it's 100% load cells, right? So BinTrac or Chore-Time or, you know, AP, or Maximus. I mean, there's a lotta ways to get load cells, BinTrac being sort of the first commercially available company that focused primarily on load cells. But now there's a million more, and I know that they're experimenting at Hanor with a couple different feed bin monitoring tools.

Smart cameras, I'm gonna talk a lot about that in this presentation. Telemedicine, I think that's pretty interesting, because he's the only one out of all of those production companies I mentioned that really is talking about telemedicine, and I'm gonna go into that a little bit more, and then robotics. The ones you've seen I think probably, if you were at the Expo you saw the robotic power washers. Robotic power washers for farrowing rooms definitely have really come into play right now. I mean, Pipestone I think has maybe 80 of 'em at this point within their sow farms. So that's just one, oh, let me just go back to that for a second.

That's one look at what one company is saying, "These are the kinds of things that we're studying." And they've got, in terms of evaluation, so accuracy, reliability, implementation, it goes back to what Dan was saying. Other people said the same thing, is that, you know, like if you're gonna put a camera in, it's gotta be up and running 98% of the time, right? So that's a big part of what they've been doing at Hanor is to look at that sort of reliability piece of the puzzle, and no surprise that in their evaluation that it's either revenue cost or profit, right? So, and that's the bottom line, although I say this all the time to people.

"Tell me what the ROI of this is," right? So that is not the way a lot of technology is chosen. A lotta technology is chosen because it makes life easier, or we can get a lot more accomplished by using the technology. Now on the feed side, what Dan was saying.

That's really well-known. Animal nutritionists, those kind of trials are very well-known how to do that kinda stuff. The technology side is a totally different picture. So at Pipestone, so Pipestone publishes quarterly, what they call the "Pipestone Journal." And so this is from winter 2021, where they went through and there was 40-some pages of, "These are the things we're studying, that we did study. Here's how we calculated the return on investment of these various things that were looked at."

You can Google "'Pipestone Journal,' winter 2021," you'll get the whole thing, PDF format, if you want to look at it in detail. But I just wanted to, I pulled out a couple things that they put in that journal. What did they look at, and what did they do? So these are the things that they looked at, but this was sort of pre-COVID, and maybe a little bit during COVID, but the infrared heat lamps, so yeah, they're gonna implement as old bulbs burn out. Farrowing heat mats, they tried 'em. They're gonna put 'em in. Robotic power washer number one, nope, didn't work.

Power washer number two, yep, implement that. I think that's the Danish one, is the one that they're implementing there. Wean pig conveyor, to eliminate pig pickup for vaccination. I'll talk a little bit more about that in a few minutes. And you can see, nope, didn't work.

And that's what Gustavo Pizarro at Pipestone will say is that the negative results are as important as the positive results, because then, we know that we don't have to even think about that going forward, right, unless there's some new versions of technology that come out there. They moved to Porcitec Mobile data entry. They use Porcitec for their sow management system. They use, they moved to that mobile app for on farm data entry.

And they did an ROI calculation on that, which is detailed in that journal, in the actual article. The ones that they're working on now, is what they said in their, in the write-up, the pig counting camera, and I'm sure that's the Ro-Main camera that many of you probably saw at the Expo. And a second wean pig conveyor that they think might be doing the trick. And, well, I'll come back to the wean pig conveyors in a minute. Processing cart efficiency, body condition cameras, and sort of a Roomba feed sweeper for the alleyways.

So these are the things that they've been trying, and among other things, right? So just to kind of give your, just to really whet your appetite about what these larger companies are trying. They're spending, like Ashley DeDecker at Smithfield said, she's shifted her focus away from sort of nutrition trials and vaccine trials to technology R&D trials and proof of concepts, and it's not that they stopped doing the other ones, but her focus has moved more towards the technology side, and it takes, it's not like running a trial on a feed additive. It's a completely different ballgame, when you're trying to do these technology trials, proof of concepts, or even implementation. So, going back to David Rosero.

He gave these two presentations. I asked him, I said, you know, "David, these presentations were great, and I wanna use a couple slides. So, can I reference your slides?" And he said, "Here, well just take my presentation, and use the slides that you want," because there's a key point to this, right? So they've been using, for the last three years, they've been using a smart camera for getting weight estimations from a company called Asimetrix, which is based in, comes out of South America, out of Colombia, South America, and they've really been doing, you know, it's almost like collaborative product development, with Asimetrix.

And so, much of the work that he is showing, in terms of the values that they've found, is coming from the work that they've done. Now, there's like five other smart camera systems for weight estimation in pigs that have emerged in the last two years, and I'll guarantee you that there will be five more in the next two years. Because people have figured out, technically, figured out how to do that, and how to do it well. Now, I think, that in most cases, what I know from the other weight estimation companies, their outcomes fall into about the same bucket, which is their accuracies, and we're talking about like, if you go out to the market weight, the accuracy gets better the heavier the pigs go. So at market, say maybe from about 200 pounds on, the accuracy is, plus or minus, about three pounds, all right? So around 97% accuracy, 98% uptime, and their take on this after doing three years worth of work is to implement really throughout their system, cameras.

But I thought this was really cool, what they've done there, which is to sort of focus on system growth monitoring, so creating a report that, essentially each row of this report is a group of pigs, right? So it's a wean-to-finish group or a finishing group. I can't remember which is which now, in terms of what they're using. I mean, this report shows finishing. But so, they're combining real-time growth, in terms of weights, that's this first section, against their targets, and then rolling in the environmental information, so high-low temps, humidity and CO2, and integrating that with mortality information, and then that last chunk, mortality is the last column. But that one, the chunk before that, is their load sell information, so they're getting real-time feed consumption, feed intake, and feed conversions.

So and they essentially then, it's just like after groups monitor and MetaFarms System, being able to look through and see what's going on across all of these groups that are on feed, and how they're performing against their targets, right? And this is one of the things that they discovered. Each bar there is a separate sort of growth check time, right, so you can see, they're checking the growth, relative to daily gain, at 110 pounds, 132, 154, 176, 198, 220, 243. This is all off the camera, right, and so they're looking at the relative average daily gain, at these points in time, against what their genetic reference, and what their target is, right? And so you can see what they found, that pattern, right, that the growth as the pigs got heavier, was a lot less than what they expected it should be, right? And so that led them to figure out, see, what's going on, and you can see it was all coming out of the Phase III weight growth period, and the Phase III diets. So they, he said, "We made nutritional adjustments." He didn't say what those actual adjustments were, but they did make the adjustments, and they did see a response.

And that, for them, that was $1.82 per pig profit, but even more, which is great, okay? That's really, I mean, if you're looking at, what's the value of putting in cameras, there you go, but more importantly, they saw this on 40% of their sites, right? So, huge, huge value. That was, that's over $2 million on their system a year.

Here's another example. So when you put these cameras in, you get this real-time growth curve. So that's, you're getting updated growth curves, smooth growth curves off of the weight estimations. You know, hourly, daily, right? So it's a real-time growth curve, and he gave, showed this example of what happened when they saw that this growth curve turned on this group of pigs, and that they intervened.

He said they used medication, intervened, didn't say what medication it was. I don't know. Maybe it was a Merck product, who knows? But they intervened, and the thing that I think is really amazing here is that those pigs turned around, and gained about the amount of weight that you would expect that if they hadn't had a problem and had their growth curve turn down a bit. And that was, for that group, that's $4.80 marginal profit. So basically, I think, what he did was say, you know, if that growth curve had continued at that plateau versus what actually happened after intervened, that was $4.82 a pig on that group of pigs. Another example, and every one of the technologies that's being tested, things like SoundTalks that BI has, the watching behavior of pigs as they, in finishing pigs, the amount of time they spend at the feeder, do they go to the feeder, how many times do they go to the feeder, or the water.

Some of the research that's been done show exactly the same thing, earlier detection than humans walking through a barn, right? So in this case, their earlier detection, you can see, it was September 20th is when the camera alerted them that there was a problem. And the first observed, that's somebody walking through the barn, first observed September 26th. So, a six-day earlier detection, and that's true, every study that's been done so far has shown the same thing. The data is a better detector, earlier, than people walking through the barn.

And people walking through the barn, looking at the pigs, is sort of considered to be the gold standard, but it's not, and it won't be in the future. It's gonna be the data that becomes the gold standard, with sort of, you know, added complementary animal husbandry, or the people walking through the barn. Finally, another example, the marketing, where they took the standard protocol at Hanor, and, in terms of observing, picking the pigs out of the pen, loading 'em up on a truck, versus following what the camera was telling them.

And what they found is that, you know, just in this case where they had some discounts, they didn't have any discounts with the room that had the camera. And they did have discounted pigs in the rooms that there was people just walking in and picking out those pigs, all right? And but even more importantly, what they saw was if they compared just their standard way of going to get the pigs and loading 'em up on the truck, versus what the camera was showing, they narrowed that weight distribution, you know, that live weight or carcass weight distribution, and they dropped the coefficient of variation from 12% to 9.7%, and then sort of mathematically, you get a little bit better average daily gain and a little bit better feed conversion on a closeout, right, because you narrowed up that distribution.

And I did some work on this. Back in 2015, this was something I published in "American Association of Swine Veterinarians," where I just showed what the, now this was on 83,000 loads going to multiple packers, about 12 million pigs. And what I did was I wanted some way of quantifying what the coefficient of variation was worth, and if you reduced it, what that brought back to you.

And that was, you could see going from 12% to 9% was about $4, $4.25 a pig, and then the real improvements, I mean, not that $4 a pig is a real improvement, but then some really substantial improvements, if you narrowed that standard deviation even more, or coefficient of variation. So basically what I'm trying to say here is, with these technologies, we're in what I'm gonna call the value discovery phase of things, right? So there's a lot of technologies that are out there that are commercially available today, or that are coming, and it's gonna be, well, it took Hanor three-plus years to come up, these are the first sort of value presentations that they've made after that work. So, I mean I think we probably have three to five years of value discovery, of these technologies, as they sort of get integrated into production systems.

Now let me just dive in to a couple things that are going on. Well, first of all, I hope many of you have seen this presentation that Andy Jakubowski from Pillen gave, about how they're using cameras, especially in their nurseries, to just look at the entire room. They've got hotel room-style nurseries, and they can put a camera in one corner or over the door, and they can get a view of the entire room, right? And they put in these simple off-the-shelf cameras that are power over ethernet, with a network video recorder, that you can log into and you can, from remote, and you can see live pictures.

Now, you can do that with a simple webcam too, but this is a continuous recording, and it stores it, right, so you don't have to, actually you can come and take a look at it tomorrow, or yesterday. I mean, if you look at yesterday, you can do it right away, in the morning, today. And the whole reason they did this is they built, they spent multi millions of dollars to build new nurseries that were then, that underperformed the old nurseries. So it was like, "Okay, let's figure out what's going on here. Let's put some cameras in."

And a couple things that were really interesting. So this is a webcam, simple webcam monitor in a corner, looking at this room of pigs. And so that's a partial fill. And you can see, pigs, pretty uncomfortable there, right? So they had higher mortality, lower daily gains, in the new nurseries than they did in the old nurseries. Also, they saw, I don't know if you can see that, vaccinations added stress. So they saw a similar picture of how the pigs looked after they were vaccinated.

I don't know if it was on arrival or a couple days later. They made a decision to move all their vaccinations back to the sow farm, because the pigs looked so bad on camera after that first vaccination in the nursery. And here's what they saw when they looked at the gruel, you know, doing a wet gruel feed. Again, pigs got wet, piled, huddling. See the pigs on the right-hand side? I mean, everybody in this room has seen all of this, over and over again, as you walk through barns.

I mean, there's nothing new here. The only new thing is, this is being looked at 24/7, via video, which is stored, and then it can be reviewed. So, one more thing that was interesting to them.

They were having these issues again with the pigs being uncomfortable, getting sick, where they noticed in looking at their environmental control data, that there were these spikes in humidity. And so what that is, they were associated with people doing chores. So you know how the pigs get when people are in there doing chores. They're running around, humidity goes up.

Well, the workers at that, in those nurseries could adjust the ventilation settings. So they saw what the humidity was, out of spec, turn up the fans, and but the humidity went right down after the chores were done, right? So then, now the pigs, now the ventilation is too high, for those pigs, and so that was also causing ill health and more mortality, et cetera. So, now, if, I just wanted to go back here for a second, because I don't. Now, it's not going to be very long before this gets automated, right? So today, they're doing all of this by hand. They're looking at all of these videos manually, right? It's not gonna be very long before somebody, or some company, and we are not working on this, but somebody will, automates this by using machine learning, because it's pretty straightforward.

You get, if you get two million pictures of pigs in nurseries, and just say, "huddling, no huddling, comfortable, no comfort," and build an algorithm around that, you can automate that and put it into software, and then that can read these videos. So it won't be very long before that becomes automated. But they're more than happy to do it this way at Pillen, and I talked to Andy about two weeks ago. He said they've got 500 cameras now in their system, not all, obviously, in nurseries.

These are the cameras and the NVR systems that they use. I asked him to give me the actual specs of what they buy. $600 to $800 per camera, installed cost.

You see what they found on their ROI was that big improvement in nursery mortality, better daily gain. But he makes the point that it's fantastic for training as well, right, and for sort of feedback for everybody working in the barns. You can power wash 'em.

The only thing that I know, Andy didn't say this, but from my experience you have to be really careful about what you wipe it with, because if you create static electricity on the lens, it just sucks the dust right back up on the lens. So in a lot of ways, if you can just let it dry, but sometimes you get the water droplets on there. But today, these cameras are pretty indestructible on being able to wash 'em down. And as he said, you can see, they focus on the first 10 days after arrival. And as I say, somebody is gonna figure out how to automate that video, and then these things are gonna rock and roll within production systems. So smart cameras for weight estimation, to go back to what, that David Rosero was showing in terms of value, these are the companies that I know about.

They've emerged in the last two years. I have personal experience with a couple of these. David Rosero's got the personal experience with Asimetrix. DOL is a Danish sensor company. I don't know anything about it except that I've seen their announcements and what they're selling. And PigBrother, sorta interesting name, is out of Hungary, and I think that one is a little bit, it's not quite as advanced as these other ones.

But, essentially, all these camera companies, and like I said, I guarantee you, in three years if I was doing this presentation again, there's gonna be five more, at least five more of these companies emerge, because now the technical stuff has been figured out, how to do it. You have to generate, if you generate three million pictures of pigs, like this, right? You put a camera up above a feeder. You generate three million pictures of pigs, and then human annotation says, and you have to weigh these pigs manually, right? That's the hard part of the development of the algorithm, but then humans say, "Oh, this pig right here, this pig weighed 200 pounds.

This pig weighed 190 pounds. This pig weighed 170 pounds." And so you hire a bunch of people to do those annotations, but how long does it take to generate three million pictures of pigs in a finishing barn? Well, probably, it's about three weeks, right? You have cameras running 24/7, to generate that number of pictures. Course it has to be by genetic line, and there's probably a few other factors that you have to figure out, but the generating of the images is not the hard part anymore, and now people know how to do that, how to build algorithms, all the statistical processing, engines are available to run those, and you can build 'em, right? So that's why I said, I think there'll be, you know, five more of these companies that emerge. This is just an example of Smart Agritech from Sweden, and sort of what you can see. I'll go into it in a little bit more detail.

Let me just check my time. How much time do I have? Five minutes. So, this is what's called unidentified individual pig weights. So each one of these dots here, anyway, you know what I'm talking about. The green line is a smooth growth curve. The dots are the individual pigs that have been weighed by the camera each day, right? So you got the fast-growing, you got the slow-growing.

That's a complete distribution of the weight of every pig in that pen, by day. And it's not like we know this is pig 1320 and this is pig, you know, 1210. But it doesn't matter to start with, right? You can use this unidentified individual pig information, just like what David Rosero was showing on the growth curve they did for Asimetrix.

And you can watch this growth curve happening. You can take the dots away if you want, the individual pig distributions, but just think about, that's 20 kilograms at the bottom, to start with, and 40 kilograms at the top, 20-kilogram difference in the individual pig weights at the start, and that holds true all the way through. So there's a couple interesting things about this. One is, how likely is it, do you think, here we go. How likely is it, do you think, that a pig here moves to up here? I can tell you, on my intuition, is 0% probability that a pig that's down here moves up here, and one that's here moves down here, unless they get sick, right? So, in other words, I think the ones that are down here are gonna stay down here, and the ones that are up top are gonna stay up top, right? So I'm gonna come back to that point in just a minute.

This is the FarmSee. FarmSee's out of Israel. That's one of their reports, but same kind of thing. You're getting the same types of reports from each one of these companies, right? So one thing, I'll put on a Merck hat for a minute, Merck has invested in LeeO, not own LeeO, but LeeO's the individual pig registration at birth.

You basically tag the pig at birth. Get the birth weight. Get lactation date and gain. Get the wean weight, using RFID tags. Traceability solution, especially in Europe. But you can, essentially you can follow these pigs all the way through.

So we, Merck, this was announced actually the day I started at Merck Animal Health Ventures, this investment in LeeO, which has really been run in the U.S. by United Animal Health and Prairie Systems. Joel Stave really has spearheaded the growth of LeeO. LeeO's used in hundreds of sites right now, or I think there's nearly a hundred customers in the U.S. for LeeO,

and growing even more in Europe. So, just think about this now. When you start to put some of this stuff together, right? Individual pig birth weight. Well, if you go back to this, here's the way I'm thinking about this. Each one of these pigs, and by the way, eventually the cameras will also be able to identify the pigs, and they'll tell you exactly which pig, and probably, technology to mark those pigs, based on a weight, right? But each one of these pigs is a cost of production. You think about that.

This is a bundle of cost of production, right, so this, We think of weaned pigs, we think of 1,000 weaned pigs as a batch of weaned pigs, and they're all the same. The cost to, the weaned pig cost coming out of a sow farm, for each of those pigs, is the same. But the attributes that they carry, that cause them to generate an individual cost of production for a pig, is different, right, and a lot of that is going to be based on their birth weight, right? So, I'd been thinking about this, and thinking, well, now which is the lowest cost of production pig in here? Is it, I thought, originally, my intuition was, well, these are the lowest cost of production pigs, right, the fastest growing ones. But, you know, maybe we would discover that that's not the case. Maybe it's the slower-growing ones, which I can't believe, but I just want to give equal time to say, we don't know yet, which is which.

But if you can predict this off of birth weight, which is where, you know, LeeO comes in, maybe we'll see a redesign of production systems. In other words, maybe what we'll do is we'll have fast barns, barns that turn over more quickly, and slow barns, right? And I don't know, I'm just saying. I'm just talking about that. Just a thought experiment, right, that if you put, if you just take, if you predict, and you are pretty good at predicting which pigs these are, and you move them into a fast barn, and they stay relatively uniform, then you're talking about a very narrow distribution when you sell those pigs, right? The same thing is true for the lower ones. Maybe you decide you're gonna sell these pigs that are predicted not to grow as fast as these other ones. I mean, no one has stopped to really consider this as a new way of thinking about pigs, and I'm not saying, identify every pig and create a cost of production for every pig.

In fact, in my conversations with David Rosero, he'd say, "20% is all you need." I know some other people in Europe that say 10% is all you need, right? You don't need to identify every pig. You need a representation of that barn. So, I'm gonna stop there, and of course, I've got more slides, and I could go on and on. But, I do think that, the analogy that I've come up with, you know, waking up in the middle of the night and thinking about this is, remember, well you know what fracking is, right? So, for oil and gas, right? So I think a lot of this technology is going to be like fracking, where all of a sudden, it's a new way, a new invention of releasing trapped value, right? So in oil and gas it was a new way to free up oil from, you know, trapped in rocks and all the rest of that. And whether you're pro or, I don't care if you're pro or con fracking.

I just am thinking of it as an analogy here. If you look at what David Rosero's work, when he was showing the value of, you know, $1.82 a pig, $4.80 a pig, $4 a pig on coefficient of variation, et cetera, et cetera, you could begin to see that there are ways to release trapped value in your production systems, right? And that's gonna take something like this new technology, and not just one of 'em, right? It's gonna have to be a combo of these technologies that we discover over the next three to five years, our sort of finding these opportunities within a production system, and I think it's gonna be a 15-20% drop in cost of production, based on these new technologies. So, I'll leave it there, and say thanks. (audience applauds)

2023-08-08 00:53

Show Video

Other news