It all started with an article we found about robot shepherds on large Australian cattle farms.

As with any new implementation of technology, the cattle farmers had a problem. According to the article, on a particularly large cattle farm the cows were too spread out for the ranchers to adequately monitor them and they were losing many animals to injury and illness as a result.

The robot shepherds used advance thermal imaging and other methods “to detect changes in body temperature and walking gait” in an effort “to improve the quality of animal health and make it easier for farmers to maintain large landscapes where animals roam free” according to the robot’s designer.

That got us thinking: what kind of data does a cow generate?

Here… have a big data apple

When you think about it, this new “Robo-Rancher” (which we would like to officially submit as the name, as the device does not yet have one picked out) is another big data technology implementation. It will no doubt be transmitting telemetric data back to a central server letting the human ranchers know exactly where on the ranch it is. In addition to it’s own data these robots will be likely be sending multiple metrics regarding ambient temperature and the cattle’s specific health indicators.

And this is just the beginning. According to the article ranching is one of the last holdouts where technology and big data specifically have not yet made any significant inroads. And this is one of the last industries on earth to not yet implement a data solution.

The same organization that is making Robo-Rancher recently designed another robot “to count every apple in an orchard so that farmers can identify areas where the yield is lower and more pollination is required.”

To data enthusiasts and scientists such as we there is nothing more exciting than seeing big data being put to such good use. We imagine the amount of time it would have taken for human workers to physically count every single apple in the orchard and then plot that information on a map or diagram with the appropriate visual language for a farmer to easily spot where she needs to increase pollination to produce maximum yield.

But with big data just such operational improvements are possible- and even required- to keep things running smoothly.

We Need Big Data

And farming is just one of those industries where we need to be running smoothly. According to a lecture given at The Ohio State University in 2014, S.A. Shearer notes that “in 40 years agricultural output will need to increase by 100%” to meet the needs of an ever-growing population.

This is quite the challenge, and it’s one that the world can’t ignore if we want to have food on our table in less than half a century. Shearer brings forth a number of options for possible solutions and turns to most notably to technology to find some answers.

His big hope? The Internet of Things and the Big Data it produces.

But how does IoT and Big Data enter into the food we eat? Easy. According to Shearer, the Internet of Things is a machine-to-machine communication network built on sensors. These sensors are constantly streaming, hence the “Big Data” that invariably follows along. He cites an NSF definition of Big Data as “large, diverse, complex, longitudinal, and/or distributed data sets generated from instruments, sensors, Internet transactions, email, video, click streams, and/or all other digital sources.” That  is quite the definition and quite the amount of data that can come from any of these sources.

He is looking at agriculture’s use of big data specifically as it relates to new equipment like precision planters that can, as they plant seeds, gather data as to the condition of the soil and the best use of land.

But in his lecture Shearer asks a very important question. “What is the value of “Big Data” if we don’t produce actionable information?”

Exactly. In his lecture Shearer showed a number of heat maps that were very valuable in showing where the land was being underutilized, but that is just the beginning. As we’ve seen with counting apples and monitoring cattle there is a lot more data out there that agriculture can generate and it needs an intuitive method of analysis if we are to understand it.

We need such a method of analysis in order to make the best use of the big data that these new farmhands are generating and improve the efficiency of our food production. If we don’t we fear a lot of empty bellies in about 45 years.

It also begs the question that if cattle ranching and corn farming have found ways to implement big data and analytics in their business to improve operational efficiency, will it really be that hard for you?

Until next time,

The VisualCrew