Big Data Meets Cattle

It is simply unquestionable that data collection is revolutionizing every single industry. This is true in raising cattle as well. As the old saying goes, you don’t know what you don’t measure, and the increasing sophistication and availability of data collection and analysis tools are making it easier to use data from farms and ranches.

Fecal Sample sheet

Because let’s face it–this version of data collection is not the best option out there.

For example, this time of year through December is the best time to check cows and heifers (female bovines that haven’t had a calf before) to see if they are pregnant. After all, a cow without a calf means wasted resources. But rather than just determining if the cow is simply pregnant or not, veterinarians can collect data to help determine why a cow is not bred. The data can also help make other decisions about how to use resources (most importantly feed and pasture) the most effectively.

Here’s how that can be done. When I ultrasound a heifer to determine if she is pregnant or not, we learn not only if she is bred, but how far along she is. At the same, we can take note of a number of other things that can be measured. This includes her body condition, which is measured on a 1 to 9 scale with one being very thin and 9 being super fat. A five is ideal. The cow’s age can be noted, as well as her frame score, the condition of her udder, her disposition, and many other things.

All this information can be a jumble if simply written on a scrap of paper or plugged into a blank Excel sheet. It has to be processed to make any sense. To help with this, I developed a program based in Excel that allows me to take the data in the field and then analyse it instantly after we’ve finished collecting it.

Below is an example of the data collection tab. The identification of the cow is entered under the red arrow. Under the blue arrow I enter her body condition, under the green arrow I enter her frame size (also scored on a 1 to 9 scale) and under the yellow arrow I enter her age in years. The numbers in the middle are the ultrasound measurements I took, while the second column on the left automatically calculates how old that makes the fetus in days. The far left column sorts the cows into three separate groups, but I can change the program to sort the cows in up to nine groups and change the names for the groups to whatever I want.

PregCheck Herd Data Screen

For those who know I’m color-challenged, I used the color reference on the computer when drawing the arrows. So unless Carolyn played a practical joke on me changed the default names for colors, they should be right. But let me know if they aren’t…because I wouldn’t put that past her.

Now once I have the data, this is where the magic happens. All that data is automatically sorted and placed into categories to show which cows were pregnant based on each criteria. For example, the program tells that 95% of the cows that were a body condition score 5 were bred, versus only 75% of the cows that were a body condition score 4 were bred. It also breaks that down into how many of each score are bred in each 21 day cycle.

PregCheck Analysis BCS

The program also does the same thing for the frame score and age data sets, as well as gives total herd information. The only thing it doesn’t do is do the preg checking for me. I’ll have to work on making a robot for that next.

With the click of a button, the program also sorts the cows in order from oldest fetus to youngest fetus. This gives the rancher a due date list in order, so that rancher can pay more attention to the cows that are likely to calve sooner and worry less about the ones that will calve later. The program also does other things, such as aggregating data from multiple herds to create data sets that encompass more than one preg checking event, to having a fillable list to make writing health papers simpler.

And if you are a veterinarian that is curious this program, yes, you can send me an email about acquiring a copy for yourself.

PregCheck Parameters Page

It comes complete with a lovely homepage that Carolyn designed, because, once again, I really can’t see colors well. And seriously, look at that picture of a handsome heifer with fall leaves behind her? What a beauty!

By organizing this data through using a program, veterinarians and ranchers can look at this data to help better allocate resources, thereby making the ranch “greener” and also more profitable. It can also help solve problems, such as why certain cows are not getting bred or are always bred late.

These questions were much harder to answer before the usage of computerized data collection and analysis. This program is only one of several out there, which cover a range of data sets from health information to the amount of feed utilized for each pound of growth. The common denominator with all these programs is they are helpful in wrapping our arms around large sets of data in order to find the correct answer to important questions.

I know that this post got a bit long and technical, but the goal here was to show two things. First, farmers and ranchers are extremely precise in using their resources efficiently and we use the finest technology available to make sure we are doing our best. Second, agriculture is much more complicated than the stereotypical “Green Acres” image that is portrayed in pop culture. Being a farmer or rancher takes brains as well as brawn to be successful. But that’s probably why people are so reluctant to leave this business, because using both to make a quality, well-enjoyed product like beef is immensely rewarding.

-Jake

 

Advertisements

One response to “Big Data Meets Cattle

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s