Thank you for joining us today for this month's University of Wisconsin Division of Extension Badger Dairy Insight webinar. A monthly webinar series offered on the third Tuesday of each month. It provides the latest research, dairybased information to improve animal welfare and breeding and genetic selection, automation, and modernization. My name is Angie Elnless, egg educator from Antois County and I will be your host today. We hope to provide this opportunity for informal discussion on today's topic. Since this

webinar format, please add any questions or comments in the Q&A button at the bottom of your screen. At the end of this webinar, uh we will it's recorded and we'll send you a link of the recorded webinar for anyone that registered. It will also be on our topic hub and also on um Dr. Victor Cabera's site. Our speakers today are Dr. Gustavo Maison and Dr. Victor Cabrera. They are both from the University of Wisconsin Madison. Dr. Gustavo is a post-doal research associate and he focuses on assessing technology adoption on dairy

farms and creates educational resources to promote AI, artificial intelligence and data technology and livestock production systems. Dr. Gabriero combines applied research, interdisciplinary approaches, and methods to deliver practical, datadriven, user-friendly, and scientific decision support tools for dairy farm management. These scientific tools are improving dairy farm profitability, environmental stewardship, and long-term sustainability. They will also give an update today on the smart farm hub which

is a collaborative research and extension initiative funded by US department of agriculture and the national institute of food and agriculture. With that take it away. Good morning everyone. Morning. So my name is Victor Cabrera. My name is Gustava Mazon and we are very glad to be here today and share about new technologies in dairy farming and update on the newly launched smart farm hub. Uh I think you're going to find a lot of nice new information here that could be very useful for dairy farming.

So before before we start uh we would like to have a little interaction with the audience and hopefully you can help us to respond this poll. Uh you can either scan the QR code on the left side of your screen or you can click the link that I believe Angie has put in the chat of the of the webinar. And what we're trying to find out here is how familiar are you with precision livestock monitoring technologies. So we'll give it a a minute or two to uh see the responses and this will update constantly and the more we do the more

uh information we get and the more we can tailor uh the talk towards the next ones and especially this link as it's going to be available online. More people are going to be able to participate, respond and provide us with feedback and that information is going to be super useful especially on the smart farm hub side to tailor how we develop uh further educational resources uh in the future. So right now we see like a even divide uh between the use of technologies with people being mostly uh a equal divide between people

familiar and not familiar. I have a couple a couple answers popping on my phone and of course after testing technology it seems that it's not working a little bit but as I can see here looks like most uh of people are slightly familiar with uh the technology. Should we refresh this to see if it uh comes up? It should have refreshed. Of course, techn we talk about technology and it not always updates as it should be as it should. Uh because now we are testing this in uh inside the the zoom and sometimes it

[Music] uh makes it different. Let's see if we have a updated version. Now we have 43 yes answers submitted are changing. That's very nice to see. So and we can see here uh very interestingly uh we have about quarter 24% somehow somewhat familiar with these technologies but also uh we have a 15% that are extremely familiar. very interesting to to to find that out as well as the fact that there is 18% of uh respondents that say that not at all familiar. So I do think what we're going to cover today it's going to be useful

to all the spectrum of familiarity with this precision technology. those who are very familiar uh will uh also find out I believe uh some new things and new concepts and ideas that could use as well as those that are not as familiar with these technologies. So uh with that I think we should move to the next uh step. So it's it's updating I guess we have no we need to go back to the presentation. I think this is all related to the to the fact that um we are in Zoom. So this is the outline. This is what we're going

to cover today and and we really are glad to be here with you sharing this. Uh we're going to talk about um a little bit introduction of the current overview of the D industry and these technologies. Um what are the technologies available? Then we're going to talk about uh interestingly adoption and perception of these technologies uh by farmers. We have some data freshly new data to share with you. And then we're going to end up with the educational resources on technology that are fully available to you uh online uh

to explore and we are going to end up with a few interesting take-home messages. So with that now I'm going to pass to Gustavo. Actually in this part he's going to talk about the precision technologies available. Yeah thank you. Thank you. So I want to start a little bit making a parallel between the use of technologies in our dairy lives and the use of technologies in cows and dairy farm in general. I think nowadays it's super common like everybody has a smartphone in their hands. Smart watches, smart

glasses are becoming even more common. And especially if you're an athlete or practice some sort of sport, there's even more more uh smart gadgets that can track your location, your performance, your heart rate, even like sometimes like blood oxygen levels as well. And in our homes is not different at all. We have all these smart devices that respond to you that we can ask about the weather. We can ask about the setting the temperature. We have cameras that do facial recognition. It's going to open

doors for you. We have we have vacuum cleaners. We have automated even lawn owners as well. And of course, the pet industry is not falling behind. We have devices that are able to weigh your cat, track the visits to the litter box. And of course, now more and more often, we're using this textbased uh artificial intelligence models to ask questions and help us even with daily tasks as well. So that brings us to like a little parallel that we're going to make to the US dair industry. If we look at the

current scenario of the dairy industry, what we see is that we're having a decline in the number of farms and then an increase on the number of cows per herd. At the same time, we're seeing a lot of labor issues, labor shortages, people having trouble finding farm work or even keeping uh farm work. And then on top of that we see increases on the total cost of production per cow per year. Besides we also have as a dairy industry a huge commitment on greenhouse neutrality. You optimize water use optimize water quality. And if we put

all of that together, we have to think that we have to become more and more efficient and more and more conscious as well about the use of resources. And one of the classic ways to increase that efficiency on the farm is to invest in technology. So that brings us to the use of like precision livestock farming. And I want to define that very clearly. that is the use of technologies, real-time monitoring technologies that are aimed at managing animals on an individual level. So for years we've been focusing

on collecting herds or even pen level data and now with these technologies we can collect data on an individual level. So on this left graph here we see that there's a huge increase on the number of publications utilizing precision livestock farming or precision livestock technologies. And here on the right we see on the blue line the use of accelerometer technologies or microphone technologies in green and even cameras um in orange. And we see that over the past decade pretty much that's been an

exponential increase on the number of studies and research being done on those technologies. And that bring us to this huge options that we have in the market. It's almost overwhelming. We have a ton of wearable sensors that we can see here on the pallet. We have different types of cameras, automated feeders, we have drones, we have scales on waterers. There's a whole combinations of technologies that can be used on dairy farm. So, one of our objectives today is just to give like a brief brush through

what's available out there, a couple of the upcoming ones, and then we'll show you actually the real data from the field. So, like I said, producers a lot of the times they feel overwhelmed by the amount of technologies. If you just look at that previous slide, there's a lot going on. So, we broke down the technologies in pretty much um four types. So, we have the parlor technologies that are mostly inline sensors. We're going to have wearable technologies, cameras, and then the robots. And we're going to walk through

each one of them, what they measure, and what are a couple management options that can be done with the use of those technologies. So, let's start first with the inline parlor technologies. these technologies I think that's been uh on the market for quite a long time and even before we talked about cameras or pedometers we already had some sort of individual measurement on the parlors measuring milk weights is not something new but now with these technologies we have a lot more data that we can use to

make decisions they're not only going to measure milk but can measure the flow of the milk they can measure composition conditions of fat, protein, fat to protein ratio, lactose and a lot of measurements of milk quality as well, conductivity, some out cell count, and milk color. And all these variables, if you track an individual animal over time, you're able to keep an idea of the baseline of the animal and see if the animal is deviating from its pattern. And with that you're able to detect any sorts of disorders and be

able to make decisions on forms either if there are changes on fat protein ratio on milk yield or even conductivity. A lot of those data they can indicate the onset of diseases like myitis or even associations with problems on the transition period for example. Now we have this big group that are wearable technologies and we're going to spend a little bit more time on those because each of those particular technologies they have different uh variables that they measure and they have different value that they can

aggregate uh for the farm and for decision making as well. So first of all we would like to define wearable technologies. Wearable technologies is any sort of sensor that's going to be fitted on the animal. So we think a lot like our smart watches um or smart glasses, they're fitted on us. So they're wearable devices. And one thing that they do, they translate the movements or the sounds into the animal behaviors. So this video here really illustrates an animal eating. And you can see by the blue ear tag here in this

animal, as the animal eats, there is movement uh on the on the animal's head. And this wearable technology is capable of sensing those movements and translating that into behaviors like eating time, rumination time, activity of the animal that can be used to make decisions on the farm. So one of the technologies that uh are available in the market right now are the pedometers. Pedometers they're pretty much one of the first wearable technologies that we had there. There's uh some data showing even work with

pedometers back in the 70s and similar to the smart watches that we have they can measure the number of steps of the animal. They measure activity, standing time, and lying time. And all of those variables, they've been used a lot for reproduction purposes as the animal tends to increase activity around heat, but also they can be used a little bit for disease detection or detecting animals that are in risk of disease. Because just like us, when the animal starts to feel sick or there's a disorder coming up, they

start to getting more lethargic. they start getting slower and they are going to lay down more and increase line time. So it's a technology that can be used both on cows and calves for animal monitoring and it's available in the market. Another technology that we saw a little bit briefly are going to be the year tags. And I think that differently from the pedometers, they're not going to be able to tell you uh if the animal is standing or lying, but just generate you an overall activity level of that

animal that can still be used to make decisions on the farm. But on the other part, they provide uh farmers with information such as eating time and rumination time. And of course, those two uh measures, they can help farmers make decisions based on dietary changes on the farm and even animal management. If feeding is happening at the same moment, there's a lot of information and value that these uh sensors can bring for nutritional management. on the farm and some uh ear tags as well. They have an opportunity

to provide farmers with the ear temperature of the animal that can also be associated with the onset uh of disease. Another very interesting uh technology that we see out there in the market are rinises and these work a little bit differently. instead of being fitted outside of the animal like a collar, a year tag or a pedometer, they're fitted inside of the animal. So, this animal is going to be fed a bolus that's going to sit uh in the rin of the animal mostly on the reticulum. And this bolus is pretty much

a lifelong technology that will stay there. It cannot be removed or tempered with which is interesting especially for animal traceability as it cannot be tempered with and it's able to measure the temperature on the room the activity of the animal as well. So similar to the other wearable technologies even it being inside of the room this technology is able to gather information on the animal activity patterns because it tracks uh temperature. It can give us some insights on the water drinking behavior

of the animal as well based on the temperature variations in the room. As the cows drink water, we have a spike down in room temperature and that can tell us a little bit of the water drinking patterns of those animals. a couple of rin bololises as well. They're able to give us room pH which is very interesting uh measurement especially to evaluate diets and make management decision based on those. And now more recently they also have a proxy for rumination. And we have to understand that the rumination measured by dispos

is a little bit different than me measured from the other technology because it's a rumination movements inside of the rin instead of the rumination that we see the animal chewing the cub. But still very valuable information that can be made either to assess uh diet management and also uh animal health as well. Now that we covered uh the wearable technologies a little bit, I want to venture into a different world which is going to be cameras. Different than the wearables, the cameras don't need to touch the animal. We can have a

single camera that monitors a group of animals and also a single camera is going to be able to record different behaviors and also perform multiple functions. So different than wearables that you need one sensor per animal and they normally do a very specific functions. cameras. A single camera can monitor many animals and record several variables that uh can help us make decisions. So to understand uh cameras, we have to present the concept a little bit of computer vision systems. And computer

vision systems is nothing but like an artificial intelligence system that is aimed to describe the world through the images. So, we're going to provide a computer in an algorithm like we see here, a normal video or a normal image like we see here, a top-down view from a group of heers. In this picture, we can see individual animals. We're able to see if the animal is on the feed bunk or not. So, we get this picture and we label a bunch of them manually on the computer and teach the computer how to

interpret these images. And the result is pretty much a human eye watching these pictures and describing as we see here on the picture on the right. The model or the artificial intelligence, the computer vision system is able to identify each individual animal. We see that the water trough has a different color here on the right. It's able to recognize the railings, the feed bunk here up top and with that provide us with valuable information on the individual level of each animal and also a little bit of the pen level as well

and what's going on. I think one of the big applications of cameras and especially one that's being applied over in the field uh a lot is cameras to measure body condition score of animals and we all know the importance of body condition score for animal management in general especially during that uh dry off and transition period and also to evaluate the negative energy balance and just manage dietary um aspects. of the animals in general. But one of the major flaw of the body condition score system is that it's very

subjective. Sometimes like cows are passing by very fast or people cannot agree within the score or there's like a lot of uh variation even on the way that the single person evaluates the same animal over time. So here we wanted to bring an example. We have two pictures. The top picture is a picture of a cow 21 days before uh calving and the bottom picture is the same cow 14 days before calving. This cows were scored by a trained observer on body condition score and this observer scored this cow as a

body condition of 4.0. But then we had at the same time c cameras evaluating the body condition of this animal. So 3D cameras that measure the distance between the camera that was placed at the exit of the parlor and uh in the aloe. So as you can see here in the picture, these cameras were able to measure the contour of the body of the animal. And then here on the bottom picture, we see that these cameras were able to detect a difference on the body shape of this animal. And this is very interesting because the cameras, the

wearables, the milk sensors, all of these technologies, one thing that they do very well is detect changes that cannot be seen by the naked eye. And I think this is a very clear example that shows uh animals that were managed visually and manually by people. And then when we incorporate technology, this technology can bring us different information than what we're seeing in real life because they're more sensitive and easier to detect changes. The computers do it uh little bit better than we do

sometimes. Another option for the use of cameras is to detect lameness. Think we don't need to say how big of a problem lameness is on farms. It not only affect affects animal health, it affects animal welfare as well. And even very similarly to body condition score, it is a little bit of subjective on how to score lameness. There are two or three different score system, a three-point score system, a fivepoint score system, and people don't always agree if an animal is lame or not. So, we have work

here that's been done at UW at Dr. Dora's lab of the development of these cameras that can be used to measure local motion score of the animals and do early detection of lameness. So for example, they can measure locomotion score similar to what we would do visually and they can also track the walking speed of the animal, stride length, head bob, back angle and a lot of other measurements that you can see each tiny point both on the top and the bottom video are being tracked by those cameras. And as we said before,

technologies they measure data on an individual level. And if we track an individual long enough, we would know if this individual is deviating from its normal. And with that, we're able to take action early to prevent the use of antibiotics or even to improve just overall animal health and welfare. So if we're able to track these animals with the use of technology for longer periods of time, we're able to act and change and adjust our managements earlier as well. Another option of the use of cameras is very similar to the first

example that I showed on the computer vision. We can use these cameras to monitor animal behavior. Here we see a study that was also conducted by Dr. Dora's group at the University of Wisconsin. These cameras were very accurate at detecting um each of these heers in the group and also in assessing the feeding time. So how many times how many how much time these heers spent in the feed bunk, the number of visits they had throughout the day, the length of these visits and the interval. So this gives us not only uh available

information on how the animals are behaving on the pan but also helps us do some assessments of a diet as well. And now more and more we're having some cameras that are able to detect some social interaction. So maybe animals that are being bullied and pushing off feed and being able to detect those as well to help us make uh action on the farm. So now that we covered cameras, I think we want to cover a little bit of like these robots, these automated systems that are a big boom in the industry. We

see a lot of people talking about them, a lot of questions about them as well. So just a brief uh touchup on those robots. I think the the automatic milking systems um it's a big thing that's happening right now especially with labor issues. We see a lot of people migrating to those or even a farm flexibility life and they can measure also a myriad of things. It's almost like you have a parlor sensor and a bunch of other technologies together. So they will measure the same variables as a parlor technology would like yield,

flow, milk composition as well. They would give you milk quality aspects as composition and color. But they also give you information on how the animal is using the system, the number of visits, the length of these visits. And one very interesting thing that they do, they feed uh the animal pellets as the animal comes. So it gives us the opportunity to provide individual concentrate feeding to the animals and better tailor better use the resources on the farm. Some um robotic milkers they even come with floor scales that

will measure body weight of the animal providing even more information that can be used for decision making on the farm. Another very interesting uh technology that we see adoption in the field as well is the automated calf feeders. I think the very classic one that we see most of the time is the one on the bottom left here on the picture for groupfed calves. So each calf will have an individual tag that will be scanned as the calf enters the system and the system will provide individual milk feeding plans to the calf. But at the

same time will also record the milk intake, drinking speed, the number of visits to this feeder. And this information tracked over time can also be used to make decisions regarding animal health as well as they tend to change if the animal is about to get sick. Another interesting option we see here on the right side is a partial body weight scale where the animals can step up when they come to do the visit and will estimate the body weight of the animal just based on the two front legs. With feeders as well, we have an

opportunity to do individual supplementations of probiotics or even electrolytes uh in some models. And now we see uh some options as well for automated feeders for individually housed cows. So these automated feeders they can be programmed to feed the animals multiple times a day and they will also record milking speed, milk intake and sometimes number of visits because the calf might not always nurse when the feeder stops by its hutch. Now that we talked about all these technologies, I wanted to brush up

really quickly on a couple of technologies to watch. I think that we have all of these very well uh implemented in the DAR industry. We see a lot of people using all of these technologies that were mentioned. But what is new? I think one thing that is very new are the large language models. And what I mean by that is this artificial intelligence that it's able to read and understand information. So it can be written information, it can be numbers and it's uses um artificial intelligence to translate that

understand and compile all this data. So a lot of applications of these can be data integration, information extraction, maybe even animal identification. We know how farm records can be confusing especially if we have a lot of transactions between farms. So doing the transcription of all those records can help us uh identify problems faster as well. One of the examples that we have it's a very clear example is the information and extraction and even simplification of information. We have a study here that

we like to use where they got 20 manuscripts. So 20 scientific manuscripts that we all know that those sometimes are have complex information or are difficult to understand and they fed into one of these uh artificial intelligence models. And after they fed all these papers, they asked a couple questions like what diseases are associated to ketosis? What uh could monsin affect? And when they went got the answer from this artificial intelligence model and these answers they rated 7.5 by experts. So a very acceptable

answer if we want to get knowledge that it's on the scientific mode and translated in real life applications. And actually we wanted to put here on the screen a couple of these tools that are available right now. If you have sometimes like an article that it's hard to understand that needs to be summarized, these tools can help you can upload articles into these tools and ask questions about it and will give you very detailed answers. Of course, we still have to be careful a little bit on how we use those. We cannot use those

indiscriminately or without any moderation and just believe all the answers. But we are going in a direction where this is very interesting especially to democratize the knowledge that we have in the science. Another very cool use of these large language models because it can understand all this complex interactions between data and words. We can use it to compile and integrate data. If you think of a dairy farm as we see here on the left side, there is data from a lot of sources. We have parlor data, you can

have nutrition data, you have animal health data, sometimes you have data that is just typed in the system as observations. So you have numbers, words, indexes all flowing in uh same pool. So these models are able to look at all that data, make sense out of it, and they're able to homogenize. So make sure that the data always sticks together and speaks pretty much the same language and integrate in a big pool. So it's even able to get information that is written and recorded differently from multiple farms into a single database.

So this is a very exciting technology that can be used not only uh right now a little bit in research settings for decision making but in the future as well to help us uh benchmark uh technologies and help us compare ourselves to other similar farms and take action that will improve not only our herd management but also profitability of our herds as well. And right now I think one of the big questions after seeing and presenting all these technologies is are these technologies really being adopted in the

field and for that we did a brief survey that now Dr. Cabrera will start touching into that. Thank you. Very good. Thanks. Thanks Gustavo. Very interesting as you you have seen. Now let's see how and and what's the level of adoption of these technologies in the field. So this is an ongoing survey. So that means still you have a chance to respond to this survey. At the end of the presentation you will have a QR code and you will be able to get to the survey if you have interest on responding. It's a very simple um 15inut

survey or less actually and the information is going to be very useful as you will see in a moment. Okay, at the moment we have 36 respondents actually uh we just found this morning that there are five more respondents which is great but we have summarized data here from 36 respondents that corresponds to more than 23,000 cows. You can see in the little table there uh the size varies quite a bit of the farms. The smallest one 15 cows very small and the largest one uh 9,000 cows. And the the medium uh size of the herds

about 260 and the average production on these farms uh around 26,000 pounds. And you see also in the map uh most of the farms as we would expect uh come from the Midwest and most of them would be actually from Wisconsin. Uh but there are some uh from from the west or northeast and south and we are expecting to have um more uh responses uh from all the different uh locations. uh this is going to be very interesting as I mentioned before to find out the adoption of these technologies and how we can better help

uh in this process. So here's one of the questions do you adopt any of the precision dairy technologies and what are the precision D technologies are basically what uh Gustavo has discussed in the previous slides in this presentation. So um you can see there very interestingly uh 80% uh four out of five uh respondents uh responded yes they do have any level of PDT precision d technologies on the farm and actually more interestingly uh the the follow-up question was how many right one two or three precision d technologies and you

can see here almost a even split uh between one two or three. It's very interesting to us to know that uh almost 38% of the respondents uh not only adopted but adopted uh three or more uh precision dairy technologies on the farms. So that gives us an indication of the level of adoption and how we can support better uh this process of precision dairy technologies on dairy farms because later we are asking what are those precision dy technologies and uh how we can move our research and extension uh uh projects to support

these uh events. Now what technologies are the farms adopting? And you can see here a split of the different uh general uh um u uh uh types of technologies on the farms. Uh starting from parlor technologies all the way to automated calf feeding. Uh once again related to the previous part of the presentation and it's very interesting to see here that 75% for example three out of four farmers indicates uh that are adopting wearables uh like uh the activity monitors for example or the rumination monitor uh uh

colors or pedometers which we all know those are very common and widely adopted uh nowadays. uh three out of four that's very interesting uh at the respondent level parlor technologies are also heavily adopted 72% uh you can see it was surprising to us to find out that uh already 11% of respondents are using some level of cameras on the farms for the decision making and there are 22% using automated milking systems and about 14% of uh levels of automated gap feeding on the farms. Once again, this is very

interesting for us to find out this level and how we can support each one of these type of different technologies on the farms. Okay. So what are the current perceptions and uh experiences of these uh uh of the use or experiences and expectations of the use of these technologies on the farms. And actually the next uh set of uh questions and answers uh are based on uh a number of statements and and we ask the respondents the level of agreement with these uh responses. Uh from strongly disagree one to a strongly

agree five and in between uh three neutral uh they don't agree or disagree. And so you can see here a summary of uh the adoption of precision d technologies uh regarding to the statement that they allow better decisions to be made on the farm and it's very clear to see in the response uh responses here uh if we see for a moment all the herds overall there is a good level of agreement with this statement. So they do believe that dairy technologies allow better decisions to be made on the farm in general. And if

we actually break out this by size of farms, you can see that the large farms have a even a higher agreement uh all the way to uh reaching a stronger agreement with this uh statement that they do believe strongly that they can do better decisions using this these technologies. But in general, in general, uh there is a strong and there is a a good level of agreement on the belief that precision the technologies are useful and allows them better decisions. Now this is another statement uh if the adoption of precision daily technologies

could improve uh service provided by the consultants. So uh once again one strongly disagree five strongly agree and you can see here once again very important and and even higher numbers meaning that there is a high level of agreement or even a strongly agreement with the fact that the farmers believe that these precision dy technologies could improve the service provided by the consult consultants. That's very interesting because uh they they may have experience or they may have the belief that uh these uh

service providers that here we're we are uh talking about consultants uh veterinarians, nutritionists could do they believe they could do a much better job if they would have some level of these precision dairy technologies installed on the farm. You can see there is some variation according to the size of the farms but in general there is a an agreement and a strong agreement in in all the cases. Next uh statement there are limited educational resources available on how to use the technologies to manage

the farm. That was the statement. Uh if they do believe there are limited educational resources or not. Uh once again one strongly disagree to five strongly agree and here there is a little of difference uh among uh size of farms. If we look first to all the hertz uh they are in the neutral uh getting close to the agreement level but u that's very interesting to see and if we contrast that actually with the large farms they are in the agreement or close to the strongly agreement. So that that for us is a very interesting finding

because we can find some difference uh among size of farms like if for example if we compare the large farms with the medium farms the medium farms are clearly in the neutral side and the large farms in the agreement or strongly agreement side. So uh here we want to dip uh further we want to dig further to find out why is that difference. We believe for example that uh maybe all the information available there is related to uh medium farms. That's normally the way we work uh and we produce information and maybe the large

farms and small farms are not um are are not in the same uh category and so they feel different about the level of uh educational resources available to them. Another reason could be for example large farms probably are higher uh level they have a higher level of adoption of these technologies and therefore maybe they find uh as they are adopting and using these technologies uh more lack of resources available out there. In any case I I think there is an opportunity to provide more educational resources for better use of all these

technologies. Now I can move on actually in a response how we can provide actually these educational resources and uh based on uh previous information and information we are collecting uh to this uh survey uh we are developing this smart farm hub uh which is a online platform to prepare farmers, students and all industry professionals for the digital agricultural agriculture ulture era uh specifically uh on on dairy and livestock production systems. So in in the bottom line of this uh platform is

democratize AI and we do that by providing knowledge and tools that are necessary to adopt and manage all these digital technologies and our team uh all the team working on this uh platform and all the information that is available on the platform uh it is portrayed here. You can see on the left side uh the faculty uh involved on this and on the right side poss it's it's a interesting and large group and and I think very um knowledgeable on what we can produce and provide as information uh as these new

technologies continue evolving. So we have a number of resources available on on this smart hub uh farmhub already there and these will continue to be up to date. We have extension articles. Uh we have a number of recorded webinars. Uh we have a list. This is going to be a very interesting a technology list available there. Farmers uh and practitioners interested on adopting or finding out more information on a specific technologies. they would be able to search the technology and find out what are the research behind

these technologies and find out to either better use these technologies or uh decide if to adopt or not these technologies for the specific goals and purposes they would have. We do have podcasts and social media most of the social media actually in Instagram at the moment and the podcasts are short videos with specific and very important messages as you can see here down in the in the screen. uh these are being released in a in a in a temporary fashion. uh we are trying to find out all the the leaders in the industry in

the digital uh livestock dairy technology and and we try to get their ideas and and their motivation and uh and and knowledge uh into the platform and obviously uh we were kind of uh put together also the system support tools available for the uh in general to to better knowledge of these technologies and adoption these technologies and better use of of these technologies and we are planning in the future as well to team up as we are doing nowadays with with extension which is great to put together courses demos and field days

and all those things that are related to the digital agriculture and the adoption of these new technologies and better use of these new technologies. And now actually uh I would like to uh move on to our next question in uh in in in in our uh presentation here today. We would like to know uh in just one or a few words what are some benefits you do perceive from adopting precision d technologies. Once again, the precision D technologies could be anyone and mostly related to what was the first part of this presentation. I do believe

we need to uh refresh this and I think it's important also to to say that well all this uh information and word cloud will be used like in the future this data will become available through the website as well and we want to share what everybody thinks about technology as well. I think uh understanding how people think and feel uh about these technologies is a huge part of like our extension mission as well. Now we should be updated in a second. So we have 11 answers flying in so far. We have a couple more

people participating and we already see um people want to know about accuracy of the technologies animal welfare being a great part of this as well think better decisions uh is showing up there more and more and I think it's very important for us as well to understand what uh everybody's thinking and of course this data is going to be used by us at the university level to perform extension u webinars, extension education as well. But it can be used by us to do research and even by technology companies or um

milk processors to understand the value of these technologies, understand what people are thinking about them out there. So we do have uh some winners there uh like uh animal welfare, better decisions, animal health. That's interesting to find that out. Uh but this is very important information as Gustavo says this is uh going to guide us in our future work better decisions. Yeah, that's that's in the center there. Yeah, a lot of uh lot of data related uh mentions as well. accurate decisions, data, precise value, accurate value, and

a little bit on the economics as well, which is a part that we still don't know a lot of these technologies. We have a couple studies out there showing uh economic benefits for reproduction, but uh we believe that these technologies, they have a much bigger impact on the management of the farm as a whole. So thank you guys for participating uh on this word cloud and I think now one thing that I will move to is just a couple take-home messages and I think one of the big take-home messages that we have on this uh webinar is that

technologies are tools they're not solutions. One thing that we have to understand all of these technologies available out there, they're to help us make uh decision needs. But we really have to understand before adopting each one of them, what are the needs of the farms? What are some bottlenecks that we need to work on? Which technology would fit best depending on each case? And of course, how to use technology. It doesn't really matter to have the correct tool if you don't know how to use it. And of course, the do technology

may cause some challenges. Uh we might have challenges with uh a lot of data coming in or what to do with this graph. New questions normally come up when we adopt these new technologies. So again, like technologies are huge management tools. They're not solutions. Don't expect to adopt a technology and have all your problems go away. They will help us, yes, make decisions on the individual level, but we have to understand that if we don't use any of the data or the information that these technologies provide us, it's

pretty much going to be a bad investment because you're not using it for what it's supposed to be used. And of course, there's some challenges that come to it. We need to change attitude a little bit on the the farm to go from this uh gut few and cowdriven knowledge a little bit to the data and cow knowledge and that's hard to do still and there are still very few actionable insights like very few technologies will give you a direct solution for a problem. they might able to point out that something wrong or

something different going on at the farm. And as even like the farmers say a little bit, there seems to be a little bit of limited resources on how to use those in general. And with that, I think we would first of all like to thank uh UW for the opportunity to present this webinar. Invite everybody to visit the smart farm hub. If you're a farmer employee, manager or consultant, please participate in our technology survey. And this QR code here will take you to a link three that will give you access to

our Instagram page, to our survey, some of our articles, and partner uh labs as well. With that, I'd like to thank you guys again and open the floor for any questions. The first question that came in um on the survey for for the kind attention um and for for uh for extension to host this nice uh webinar. Uh I believe this is going to be recorded but uh we will be more than glad to entertain some questions or comments about the today's presentation. Yes. Uh we have a couple questions that came in. Um I think you are muted. Uh an

can you hear? Hello? Yes, we can hear you. Can you hear right now? Yes. Uh the question came in on the survey. What is the sizes of the small, medium and large farms? Uh the herd size on the small farms was less than 95 cows and the large farms I think they were more than 600 cows. So the large range from uh the average size of the large farms actually was 2500 if I'm not mistaken. All right. uh an anonymous attendee thanked you for the excellent summary and they wanted to know what your exact percentage in numbers of the savings has

been reported with using both AMS and calf feeding uh robots. you mentioned that in your uh part with the robots uh automatic calf feeding if I'm not mistaken 20% uh close to 20% of the respondents were using automatic calf feeders and that's uh interesting data especially if we think that 60 to 70% of the calves in the US are individually housed so it shows that uh whoever does not individually house their calves s might be a large adoption of that type of feeder. The robot feeder is probably close to 10 15% if I'm not mistaken as

well. And then do you know specifically any labor saving percentages on those the AMS and the milking robots? Uh not on the top of my head. I know there's uh some literature out there that I'm more than happy to look and share with extension to be put like uh on this webinar later when it's published. All right. Are there any other questions from the group? We'll leave uh I hope everyone had a chance to scan this QR code and the survey. Please fill it out. everyone is um available to fill it out and then we

would like to just have a couple announcements here quick in the last minute before we close out. I hope you enjoyed the webinar today and uh next month we are changing it up a little bit for the Badger Insight. We're going to have a producer panel on focusing on four dairies reproductive strategies for high fertility results and the panel will be moderated by Dr. Paul Fricke. So it's going to be an interactive panel. Uh joining him will be Dave Joi, Mike Martin um for Maple Ridge Dairy, Jeff McNeely, and Chuck Rip. And to register

and learn more about next month's webinar, please visit gowisk.edu Badger Dairy. And then for additional information or unbiased university-based resources, please visit any one of those um sites that you see up in your screen. But uh thank you