National Hog Farmer is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Photo of a hand holding a smart phone Michigan State University

Helping animals, helping farmers

Data and technology offer ways to improve quality of life for farmers and animals.

By Alex Tekip, Michigan State University
For the past decade, Janice Siegford has been investigating ways that data and technology can be used to improve animal welfare practices and allow farmers and producers to better allocate their resources.

A professor of animal behavior and welfare in the Michigan State University Department of Animal Science, Siegford is quick to emphasize that she doesn’t view data and technology as replacements for farmers but rather as a tool that enhances their capabilities.

“I don’t see the technology as a substitution for humans — I see it as letting us do the things that we’re better at doing,” she says.

With fewer workers on farms, Siegford says technology is becoming even more essential.

“Sometimes there’s one stock person for 100 cows, or 500 cows, or 200,000 chickens, and so how do we know what’s happening on a daily basis?” she says. “How do we know what their quality of life is like if we have so little interaction with them?”

Technology can play an important role by providing insight into what the animals are experiencing.

Siegford’s first foray into working with technology was at the MSU Kellogg Biological Station farm, where she was part of a research team studying how cows adapted to being milked by robots rather than humans.

“With robotic milking, a cow has much more choice over when she milks based on her wants or her needs,” she says.

Each cow at KBS had an individual tag that fed information into a computer about how much the cow milked, how much she ate, her weight and other traits. Each cow was linked to a specific dataset.

“There’s so much information we can get about whether this cow is following her normal routine or whether something changes,” Siegford says. “Maybe she reliably comes every day at 8 a.m. and 8 p.m., and then one day she doesn’t. Why doesn’t she?”

Data like this provide a way to easily identify and investigate changes in behavior.

“If you’re getting constant data from animals like those cows in the milker, then you have a much better sense that things are either going really well all the time on this farm or maybe not so well,” Siegford says.

Finding the best way to analyze data can be a challenge, especially when there’s so much of it.

“We have an ability now to look at animals’ lives individually on an almost continuous basis,” Siegford says. “But it’s a lot of data. It’s massive.”

Pigs: Automatic feeders and computer algorithms
Pigs are Siegford’s primary animal of study now. She is collaborating with fellow MSU animal science faculty members Juan Steibel and Cathy Ernst. One of their projects involves working with animal behavior researchers from Scotland’s Rural College on automatic feeders.

Each pig is given a unique radio frequency identification tag, which the automatic feeder recognizes, and when a pig goes into the feeder, it records how much it weighs, how much it eats and how long it stays. Some pigs are dominant enough to eat whenever they want without being bothered: others get picked on and followed and have to sneak in during off-times to eat.

“You can learn a lot about the social structure of the group just by looking at which pig is feeding when and whether it gets to feed without interruption,” Siegford says.

Siegford is also using technology to examine pig aggression in production environments, where the experience is much different from the natural family dynamic.

“It’s kind of a ‘Lord of the Flies’ situation,” she says. “We have all these young pigs that get weaned all at the same time and sorted into same-weight, same-sex groups. This is the perfect storm for competition because, when all the pigs weigh the same and are the same gender, they’re so evenly matched that when they fight to establish their social rank, they fight really hard to sort it out.”

Siegford is also collaborating with researchers at the University of Leuven in Belgium to use computer algorithms to analyze video of pig aggression in group housing.

“We use 3-D cameras so we can subtract all of the background from the pigs and their shapes,” Siegford says. “You get certain kinds of degrees of motion and the way the vectors move that you can distinguish from one pig scampering through the pen and playing, two pigs fighting, or two pigs lying next to each other and being peaceful. We kind of make these distinctions between how much area is occupied and what’s the speed and intensity of our angle of the movement.”

Identifying aggression and pigs can be used to better manage pens, including feeding; and select pigs for breeding that are better suited for group housing.

Chickens: Accelerometers
An accelerometer — essentially a “chicken Fitbit” strapped to an animal’s back — is used to gather data on poultry behavior. Subir Biswas, a professor in the MSU College of Engineering, was key partner with Siegford on a project using accelerometers.

“We were working to recognize the movement signature of the accelerometer to tell when a chicken was standing, or sitting, or feeding, or drinking,” Siegford says. “We could actually identify specific behaviors based on the movement signature of the accelerometer.”

Identifying the performance of specific behaviors could help farmers determine if chickens are performing normal behaviors in right amounts, or whether changes that could indicate an issue like sickness or feather pecking have occurred, according to Siegford.

“Sick birds eat less and walk less — changes that could clearly be detected by sensors that recognize these behaviors,” she says.

The accelerometer measured movement in three axes: horizontal sideways movement (x axis), vertical movement (y axis) and horizontal forward movement (z axis).

“When a chicken walks, you’ve got the forward motion, but you’ve also got a little bit of side-to-side jiggle as it takes its steps, and there’s a little bit of body bobbing,” Siegford says. “You get a little bit of vertical, a little bit of body shaking as they’re mostly moving forward.

“You have these other key characteristic types of head bob and little side-to-side motions that occur with walking that are very different than when a bird is standing still and pecking,” she says. “You can distinguish between those changes in the x, y and z axes in terms of how much motion each one is showing.”

Although the project ended in 2015, Siegford is currently exploring funding opportunities to continue related research.

Farmers and producers: Increasing efficiency
Using data and technology to analyze animal behavior not only helps improve animal production but also helps farmers and producers use their time more efficiently.

“You can use the technology to help you know what the animals are doing and where you have problems that you need to address, and you can target the humans in a smart way,” Siegford says. “We can have the humans dealing with the things that require human-level skill and intelligence rather than doing low-level grunt work.”

Siegford mentions dairy farms as an example.

“If you have a robot that is now milking the cows, you can look at the alert list and say, ‘Here are the five cows that need my attention out of the 500,’” she says. “You can spend time actually watching a cow, interacting with her and treating her rather than spending it cleaning udders and sticking teat attachments on.”

She says data and technology allow farmers to maintain an environment of positive animal welfare.

“Data now tells you how to do the things you wanted to do as a farmer, which was to take care of your animals better and to interact with them most appropriately,” Siegford says.

This article was published in Futures, a magazine produced twice each year by Michigan State University AgBioResearch. Click here to view past issues of Futures. For more information, contact Holly Whetstone, editor, at 517-355-0123.

Source: Alex Tekip, Michigan State University, who is solely responsible for the information provided, and wholly owns the information. Informa Business Media and all its subsidiaries are not responsible for any of the content contained in this information asset.
Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish