Automation and Strategic Harvest Technologies

Jackson: Welcome to the FarmBits Podcast: a product of Nebraska Extension Digital Agriculture. I'm Jackson Stansell,

Samantha: and I'm Samantha Teten,

Jackson: and we come to you each week to discuss The trends, the realities, and the value of digital agriculture.

Samantha: Through interviews and panels with experts, producers, and innovators from all sectors of digital technology, we hope that you step away from each episode with new practical knowledge of digital agriculture technology.

Jackson: On this second full episode of the FarmBits episode, we are pleased to welcome Dr. John Evans, assistant professor of Agricultural and Biological Engineering at Purdue University.

Samantha: John got both his bachelor's and master's degrees from the University of Kentucky in Biosystems and Agricultural Engineering, and his Ph.D. in Biological Systems Engineering from the University of Nebraska - Lincoln. While at the University of Nebraska he met his wife, Rachel, who is from southeast Nebraska, and they now live in West Lafayette, Indiana, with their dog Maggie.

Jackson: His research interests primarily center around machine logistics optimization, machinery automation, data acquisition, and data management and analytics. All of these research topics pertain directly to harvest operations, particularly future innovations expected in harvest operations.

Samantha: Farmers understand the logistics challenges that the harvest season presents, and they also understand the importance of measuring crop productivity in the field.

Jackson: Today, we will talk with Dr. Evans about specifically these topics - harvest logistics and yield mapping - to gain an understanding of contemporary technologies and to glean some insight into opportunities for innovation that may make farmers lives easier during harvest.

Samantha: Here's our interview with Dr. John

Evans: 

So, John what do you think are some of the

growers greatest challenges during the harvest season?

John: Oh, I think it's that coordination of machinery and labor. So, trying to make sure that your very expensive machines are you're not putting more hours on those than you need to, but also the coordination of labor. You know often it's seasonal labor and how do you get in there and make sure that they're productive and not sitting around waiting on you to get done with things. And that's a big challenge and there are so many variables, it's pretty daunting.

Jackson: Sure, can you speak to a few of those variables that a farmer might need to consider when they're planning those harvest logistics around harvest time?

John: You know, a few of them are going to be your actual machinery needs, so you need to match your labor to your machinery which is often a whole other challenge in itself, but then the timing around that. So, when is that labor going to be there and is it going to coincide well with when you actually need to have it there. So, often weather is a big factor in that, especially around soybeans. You don't ever know, it's hard to know sometimes when soybeans are going to be ready to cut depending on the level of dew or humidity it could be 8 AM or it could be 2 PM so how do you manage that? And once you start going, how do you manage some of the outside factors like wait lines at the elevator? You know, that's stuff that you can't predict but you have to deal with, and if you aren't nimble and you don't have a back-up plan you can end up sitting and waiting for a long time, and you want to try and avoid that as much as possible.

Samantha: So, with these challenges and you mentioned having a back-up plan, how much time do you think growers should spend on planning their harvest logistics?

John: I don't know if there's a specific number but I think a lot of it starts even before planting. Most people are already doing this but if they're not they need to think about it is-trying to time up the maturity of their crops so that everything is not ready to go at once because then you start getting into field losses which we also want to avoid. But then if you go through the season getting up to pre-harvest there's a lot of work in terms of you had a plan in the spring but depending how the summer went that may not be viable anymore. Going through doing scouting and looking and seeing when you think each field is actually going to be ready, having a plan for the order of those fields, and it may not actually work out, but you need to have some idea going in. And I would say a lot of producers especially owner-operators who are in the combine spend a lot of time thinking about harvest logistics in the cab. I know a lot of the ones I've worked with, they're constantly thinking about that in addition to all the other things they have going on with trying to manage the harvester itself. And so, like I said, I don't know if there's a specific number but there's a lot of time that should go into it.

Jackson: Sure, and you kind of already spoke to this in terms of soybeans you know and different soybean maturities and how weather and field conditions can affect that harvest logistics planning, I guess looking more into and considering in the context of the derecho that came through in Iowa this year for example, how can some of those field conditions that are affected by weather impact how you go about planning your harvest logistics? And maybe getting some of those still standing crops before getting into your more lodged areas, stuff like that?

John: Yeah so it can be quite challenging. Like you said, I already mentioned soybeans,

you know the variability in when they're actually going to be ready to run and even when you

have to stop, but another good example of that like you mentioned is if you have downed corn planning for that. That's going to take a lot longer to get through, usually requires some more equipment, so that's a whole other challenge in addition to soil versus crop conditions. Your crop may be ready to run but in areas, like if you have a bottom field that has poorly drained soils, you probably shouldn't be on that because you're going to do more damage in the long run. Being able to work around that, so maybe that field was in the plan but maybe you skipped that one to go on and do a field that's actually ready from a soil and crop condition perspective. There's just a lot to think about.

Samantha: So, can you speak to why this is so important, what are the potential savings associated with having a really efficient harvest plan versus maybe one that's not so efficient?

John: The two obvious ones are going to be your labor cost and your machinery cost. You obviously don't want to pay for more labor than you need because you have bottlenecks

in your system and people are just sitting around doing nothing. At the same time, you don't want to put more hours on your machinery but probably the biggest one is going to be your field losses. If you're not getting in and out in an appropriate amount of time and you get pushed back into late November or early December, then your chances of a weather event coming through

that are going to increase your field losses really starts to get more and more. We always worry about wind with corn especially when it gets brittle and you know we worry about snow with soybeans taking them down, so, we need to be as efficient as possible getting in and out of the fields.

Samantha: What do you think is the best thing a grower can do today to help them improve their harvest operation?

John: I think there's the simple approach with just being analytical about identifying bottlenecks. That's something that doesn't really take any money, it just takes a little bit of time in terms of figuring out what's really taking the longest. Is that you're constantly waiting on trucks because you don't have enough or because your elevator is slow and you spend hours in line

there? So really trying to identify those pinch points and looking at ways to alleviate those.

That's something that farmers can start out with and a lot of them are doing that already. But then, you need to take an extra step to say, ok lets start taking a log of how much time I actually spent waiting on the truck to get back. Then you can really start putting numbers to the evidence you've seen in terms of how much time it's costing you in the field. And then you can start looking at ok well I can afford to get another truck if it saves me x amount of time. So, that's kind of the base approach, but then you can start, you know as you get more into that, you can start getting into some of these new offerings from different companies. But, I think in the next few years, we are going to see a lot more coming out from the bigger companies.

Jackson: So, are there any good softwares or tools out there that can help farmers make decisions about you know how to be more efficient? Can they evaluate some of those savings or is that something we are still waiting on technology-wise?

John: We are starting to see some of that. We've got companies like CropZilla that are working on trying to help farmers make harvest plans or just whole farm plans really. And then from the, and that's really more on the order of fields you are going to go after. And then terms of actual in field logistics, you are starting to see more companies look at that. A lot of big three ag companies are looking at that. They haven't really rolled out anything to date but there are some smaller companies like Solinimac, for in field logistics. And I would say, one thing that is yet to

really be harnessed, at least widespread, is the power of IoT and connected machines. So, you know, that's kind of a color barrier because most manufacturers are pretty good at handling their own machines, but most people don't have a fleet of purely green machines, especially when you talk about the whole system. John Deere doesn't make a semi-truck. And so, they aren't really good at handling that aspect which is really a big one. And so, I think that is going to be really

big going forward is connected machinery and building some intelligence around that. Help the farmers so he is not sitting there spending a lot of his time thinking through these harvest logistics while he is in the cab and getting some help there is going to be big in the future.

Jackson: So, kind of synthesizing a few of your answers, of the last few questions, you have talked a little bit about how there is an opportunity for IoT and connected machines and you know trying to communicate across machines. I know Sam and I have talked to growers who have said that their biggest issue is trying to figure out if they should send the truck to town with grain or if they should sendit directly to the grain bin.

You know, and obviously data communication across machines is a huge issue trying to make this all work. So, growers are kind of in a situation right now where their piece-mealing it between several different apps where they have just a regular tracking app that they are trying to use to see where the truck is in relation to the field they are harvesting at. Is there an opportunity here for a third-party company to step in and you know, operate off the CANBUS systems on these different machines and you know, try to coordinate them and try to get this into an application? Or what are your thoughts there?

John: Yeah, I think so and I have seen it happen one time with FarmMobile. They have, I have seen a planter-based system where they have been able to put together. I think the biggest issue right now with them is, at least from the growers' perspective, is the cost of the equipment and then the, at least right now, all they are getting back is essentially, the positional data in terms of, I mean they might get some information from their grain cart, but to really make this system more powerful, we need a kind of a "brain" around it. So, that it can essentially, you this is another set of data that producer right now has to ingest and turn into information, you know. So, if we can take some of that information burden off of them and say integrate some stuff from Google, or any kind of mapping software, and say "okay, my truck is here", and tell the operator based on today’s operations, we think your truck is going to be back here in 30 minutes or he is already on his way

back and we know that it should take him 10 minutes to get back. And really start to turn more of that data into information for the farmer. I think there is a big opportunity there. You look at Google right now and you pull up a business and see when they are usually busy, you know getting that kind of technology to the grain bin or you know, to the elevator level would be helpful. I don't know how we get there, but that is one that continually, it is such a big piece

of the pie that producers do not have any control over. The more we can do to help solve that, and at least give them insight into what is going on there, the better off we are going to be. And I don't know how you do it, because most elevators are not going to be very forthcoming of that, because they don't want to drive business away from them, and so, the way Google does it with businesses is just huge numbers and you are not going to have that with elevators or most elevators. So, I guess I think that is one thing that is worth looking into is how to get past that hurdle. As you said, it is so big for planning um and also in terms of, I think that also justifies a lot more on-farm storage. So, you know if you have enough farm storage, you can start using that as a buffer and you don't have to take anything to an elevator until you have a rainy day and you are just pulling it out of your own bins. I have seen people do that where, that way they know what their system is going to look like.

Samantha: That is really interesting. It is also really interesting like different regions of the US have different amounts of on-farm storage. Like I think Nebraska has a lot of on-farm storage.

John: Yeah, you know, especially around. You know it is also interesting regionally how the density of elevators. So, you know, where my wife is from in southeast NE, there's a quite a few small elevators that they can pick and choose, but you know, I kind of thought it would be similar here in West Lafayette, but talking to the Purdue farm managers, there's really not many options. And they are even consolidating even more because of you know, OSHA regulations, like the companies, like ADM, don't want to spend a lot of money on small elevators so they consolidate more and now guys are having to truck 40-45 minutes and don't know what the line is going to be. So, it gets really prohibitive and really incentivizes a lot of on-farm storage.

Samantha: That is interesting. So, switching gears a little bit, where do you think harvest operations are heading with increased automation? Like will we get to the point of driverless grain carts or combines?

John: So, I think we are already, with companies like SmartAg, you already have some demos of driverless grain carts. Again, a lot of the big ag companies are working on this. I think we will get there. I think grain carts will be the first domino to fall. Combines are going to be a lot trickier. We are already heading in that way in terms of sub-system automation. So, if you look at most of the big machinery companies offer some kind of process automation. They are looking at grain quality and then doing the adjustments to the combine automatically now. So that is a big step. I think the actual autonomous navigation is not too far off. I think there is a lot of work to do in terms of machine vision. We are at a very complicated environment. We are trying to look through corn, look through beans, to try to identify things that are not supposed to be there. And so, I think there is a lot of work on that and then on the legal side of it is, you know these are right now very big machines and they are only getting bigger. And so, at what point do you trust them to not get loose, because if they do, they are going to cause a lot of damage. I think we will get there.

Jackson: So, you talked about, you just talked about how machines are getting bigger, and I guess we have also seen some of this trend in the automated machinery space to move towards smaller tractors. Do you think that smaller autonomous vehicles, or kind of swarm technology, will ever have a place, specifically in harvest operations?

John: That is a good question, I think, harvest operations are going to be one of the tougher

spaces to go smaller. Just because of the cost of technology and process. It really is an economy of scale there that is working in the machines favor to get bigger. In other operations, it will certainly happen faster. I think there is an opportunity, it is just going to be as the cost of technology gets less and less and I don't know how long that is going to take. But when you think about all of the sensors you are going to need in one of these big machines, that is just going to multiply when you get into swarms, so if you want good GPS accuracy, you need to have RTK, and that is obviously not cheap. And then you know all of these other machine vision components, you know, get cost prohibitive to go smaller. Right now at least.

Jackson: Considering that same theme, you know of being smaller, I have seen that some of the companies are kind of moving towards trying to get into row by row resolution. I know last year, we had some data come in from one of the new combines, it was coming in at around 5 hertz, and they were reporting at you know, sub full-swath width resolution with their yield. And so, kind of getting into this idea of yield mapping and yield data resolution, how much resolution do you really think we can really get to for yield data without much error and what would be the technology pathways we may take in order to get there?

John: I think we can probably get, I think it's crop dependent. I think corn is going to be easier and I know there are companies that are looking at using machine vision systems to try to get a distribution of the grain coming in and then just saying we know what we got from the yield monitor, this is how much volume we have and from the vision system we know the distribution of the volume. I don't know right now how accurate that's going to end up being but for that we could definitely get down to the per row basis. Soybeans get a lot harder. It's really hard to tell that from a vision system especially with the volume of plant coming in. So, I think we will potentially get there, and soybeans are also hard too because there maybe, the platforms are also often much wider and so you really do need to get better resolution but it is like I said a much harder problem than corn. In terms of value, I think there is a lot of value especially when we're using yield data maps to make prescriptions for fertilizer and other inputs. Right now, in terms of planting or applying fertilizer we can get really spatially dense, I mean easily down to the row and along the row as well. So, I think it'll help, I think there's still a lot of research to be done in terms of we're also adding in a lot of other data layers like soil fertility that are still very sparse and that are cost prohibitive to go much denser. So, that may end up being the limiting factor anyway. It's just, I think there's a lot more research to be done around that area.

Samantha: So, to leave you with this, what is the biggest piece of advice or message that you want to leave the listeners with?

John: I think it's just to continue to try to get better every year. You know, you don't have to go out and buy the latest and greatest right now, even incremental change is going to help your operation. Like I said, farming is pretty unique in that, especially in row crop grain, all of your

income comes at the end of the year and how you handle that harvest can have a big impact on that. So, if you can continue to get better, whether that's through just manually tracking where

your bottlenecks are and trying to resolve those or it's incorporating some of this new technology, it's all going to help you in the end. So, just continue to work and get better at

your harvest efficiencies.

Samantha: Thank you Dr. Evans for joining us today. John was actually a former member of our research lab, so it was great to catch up with him and see how he's doing and the things he's working on now! I thought he did a great job covering things that growers can do or use today but also where we're going in the future. So, my favorite part was when he talked about how most people do think about their bottlenecks or things slowing them down, but maybe don't actually write down a time that they're spending on that to then justify some big purchase decisions.

Jackson: Exactly, and what was really interesting to me, when John talked about the logistics issues of the trucks communicating with actually the field machinery and trying to understand

wait times at the grain elevator, and trying to make decisions about "Do I take it to on-farm

storage, do I take it to the grain elevator in town?" And where we could potentially go with that,

maybe using technology like what Google, Waze, what they already have out there - that's

really cool stuff.

(Yeah it is!)

And I'm excited to see what research goes on there in the coming years. So that's all for today's episode, I really hope that you'll join us next week as we talk to Dr. Joe Luck about yield data quality and also yield monitor calibration and yield monitor function.

Samantha: Thank you for taking the time to

join us today on the FarmBits podcast.

Jackson: We would like to thank Nebraska Extension

for their support of this podcast and their commitment to providing high quality informational

material to members of the agricultural community in Nebraska and beyond.

Samantha: If you enjoyed this episode and it sounds like something you'd listen to each week,

subscribe to the podcast and set your notifications to let you know each time we release a podcast.

Jackson: We would love to hear from you with

your feedback, so if you have comments or questions for us,

please reach out to us over email at NEdigitalag@unl.edu,

on twitter @NEdigitalag, or in the reviews

section of your favorite podcast platform.

Samantha: See you next week on another episode

of FarmBits.