When someone thinks about data analytics, usually it’s coupled with images of researchers sitting in front of computers, number-crunching using sophisticated software amid a futuristic setting.
Cows standing in a field, on the other hand, do not leap to mind.
But the pastoral setting down on the farm has provided a great example of how data analytics can improve business in all types of industries.
Brandon Perkins, a product manager with GE Industrial Intelligence Data, told the Farm Journal he began to see the potential several years ago after attending a conference in which he learned about the use of predictive analytics to detect illnesses in cows, which had a better record than cowboys using their eyes, experience and intuition.
Turns out the cowboys were taking out too many cows that weren’t sick, wasting time and money in “servicing an asset” that didn’t need servicing. Analysts used data gathered from sensors placed on the cows that monitored movement. They eventually found patterns in certain types of movement and began to identify behavior that indicated illness.
Their predictions proved more accurate than the cowboys’ observations. It also showed data is “the fuel of the 21st century,” Perkins said.
The Internet of Cows
In April 2017, the Brazilian start-up company BovControl held a hackathon in San Francisco that gathered programmers from across all industries to help solve the challenges surrounding the use of data to improve cattle farming.
BovControl’s stated goal is to “connect every cow on the planet to the cloud.” That will lead to better data collection and enhance analysts’ ability to provide information to farmers that will improve cattle operations.
The event, which also included the involvement of Google Launch Pad and Silicon Valley Forum, involved teams which looked into solutions for challenges faced by cattle farmers. Those included ideas for small ranchers to secure inventory financing, a tracking platform for lost cows and a web app designed to connect organic cattle farmers with exporters interested in their product.
Japanese Dairy Farmers
One of the best examples of data analytics and the cattle industry comes from a book that was a bestseller in 2016.
In “Thank You For Being Late: An Optimist’s Guide to Thriving In The Age of Accelerations,” Thomas Friedman wrote about an interview with Joseph Sirosh, vice president of data in Microsoft’s Cloud and Enterprise Division.
To explain the many applications of sensors and data analysis, Sirosh used the example of cows.
Dairy farmers in Japan had contacted Japanese computer company Fujitsu and asked if there was any way they could help to improve the success rate in breeding cattle on their farms. The main problem: Cows only go into heat for about 16 hours every 21 days, mostly at night, leaving farmers to guess the best time to artificially inseminate them.
After researching the issue, Fujitsu engineers decided to equip each cow with a pedometer that transmitted information on their movements back to the farm using Microsoft’s cloud technology (called Microsoft Azure).
It wasn’t long before the data showed a trend that no one knew about – female cows in heat take faster and more frequent steps. Once the researchers figured this out, the system was programmed to send a text alert to the farmers when the steps increased.
Just like that, the farmers could stop guessing and use the data to have better success rates in breeding their herds. Sirosh said the farmers experienced a “huge improvement in conception rates.”
The work in Japan also yielded other insights into cattle.
- Data showed that a cow inseminated in the first four hours of the 16-hour window for artificial insemination had a 70 percent probability of giving birth to a female
- Insemination done in the second four hours had a higher percentage chance of producing a male
- By looking at movement data, researchers were able to find footstep patterns that offered early signs of eight different cow diseases.
As sensors and methods for crunching statistics becomes more sophisticated and farmers become more aware of the potential of big data, expect data analysis to create more ways for ranchers to improve their cattle operations.