Data Revolution's Human Touch

Human Intuition Still Needed to Make Complex Cropping Decisions

Todd Neeley
By  Todd Neeley , DTN Staff Reporter
Connect with Todd:
Brandon Bell shares data from his operation with Co-Alliance, which then develops variable-rate planting maps. (Progressive Farmer photo by Dave Charrlin)

OMAHA (DTN) -- Chalmers, Indiana, farmer Brandon Bell isn't waiting for that digital ah-ha moment when, with a simple mouse click or two, a computer generates game-changing recommendations that match the complexities of farming.

To get top recommendations now, Bell shares soil and yield data with his local soil-fertility experts through a local agribusiness and energy marketing and supply cooperative. The data form the basis for his variable-rate fertilizer applications. Sharing crop information allows him to benefit from the data he collects now instead of holding out for the development of an all-knowing algorithm to improve decisions on his 3,200-acre farm.

"Everybody overcollects the data," Bell said. "You tend to sort through what you need and what you don't need. I'd say every person is different. There are definitely guys out there who will spend as much time on data as anything. We're 25% planning and using the information. But, 75% of the time, you gotta work and make it happen."

Bell primarily runs John Deere equipment and links to a mobile app on an iPad on his corn and soybean farm, north of West Lafayette. He shares his data with Co-Alliance, which serves as a go-between for the farmer and a fertilizer plant manager. Editor's Note: DTN has a business relationship with Co-Alliance.

"It has been a big savings," Bell pointed out. "It allows Co-Alliance to build variable-rate planting maps. As far as the variable rate, we've seen cost savings and yield increase, and we've seen increases on bigger-yield land."

BROKEN PROMISES

There has been no shortage of companies promising big returns to farmers who collect and supply data on yield, soils, financials and other details. Just put the data in and algorithms will spin out the perfect answers, the promise went.

"That turned out to be really, really wrong," explained Bruce Erickson, agronomy education distance and outreach director at Purdue University.

Along the way, unpredictable real-world challenges arose. Every field is different, and every farm is different in the crop-production challenges it faces, issues an algorithm can't anticipate, at least at this point in the data evolution.

P[L1] D[0x0] M[300x250] OOP[F] ADUNIT[] T[]

Sharing data with a local agronomist or seed dealer and crafting a plan of action on the farm is likely something that will always be needed.

"I wondered if the age of agronomy was over," Erickson explained, thinking back over the past 15 years. "It seemed to be the business aspect was more important," and computers would supply the crop-production answers.

Now, agronomists are demanded more than ever.

COMPUTER, HUMAN COOPERATION

In a 2012 TED Talks, Shyam Sankar, director of San Francisco-based Palantir Technologies, said human/computer cooperation would make the use of data more effective (visit www.ted.com/talks/shyam_sankar_the_rise_of_human_computer_cooperation?language=en#t-707842).

"Isn't supposed to be man versus machine," he said during the talk. "Instead, it's about cooperation and the right type of cooperation. So, if you want to improve human-computer symbiosis, what can you do? You can start by designing the human into the process. Instead of thinking about what a computer will do to solve the problem, design the solution around what the human will do, as well."

Palantir builds intelligent operating systems software for government agencies and commercial industries, with an eye on agriculture.

Ted Mabrey, head of business development at Palantir, said he believes artificial intelligence has the potential to improve decision-making on the farm. Human intuition, however, still will be needed to maximize the collaboration.

"The primary problem with technology is finding a way to manage complex data that can be interpreted by someone who is not a data scientist," Mabrey explained. "Who can interpret data to take to the farmer and empower them as the decision-maker?"

The data-technology problem yet to be solved, he adds, is meeting farmers where they are. He said it has to start with talking to farmers about the decisions they are making. "Technologies that really work start with operators," he continued.

NATURE OF THE BUSINESS

Alexander Reichert, CEO and cofounder of AgVend, a digital market for farming inputs, stresses the data revolution hasn't arrived in agriculture because of the complexities of the business.

"There is so much variability in this industry," he said. "You can farm two different tracts across the street from each other and get two different results. That makes it very difficult."

Purdue's Erickson notes the volume and use of data in agriculture will continue to expand but will not be fully realized until technology catches up to the industry.

"In science, you can mix chemicals and know what will happen," Erickson pointed out. "In a biological system, you are dealing with living things, soil, plants, etc. -- complicated and mysterious things. The simple things are more complicated than we thought. Even fields that don't look complicated are complicated."

Bell said he's not opposed to sharing data regionally to better other producers. After all, it's common in northwest Indiana to produce 240-bushel corn yields, he added, so producers really aren't giving up any secrets.

"As long as it is for the betterment for what we're doing, I don't see a problem," Bell said. "As it happens, things need to be simplified for farmers to be interested. There's nothing more frustrating than sitting and waiting for maps to download.

"There will always need to be someone involved locally," Bell continued. "We're going to have automated tractors, and they will work. But, I'll not be able to sit on a beach in Mexico while it's working. Technology and data is great, but I don't know that we're ever going to get away from being physically involved. I still want to be involved."

Todd Neeley can be reached at todd.neeley@dtn.com

(AG/SK)

P[] D[728x170] M[320x75] OOP[F] ADUNIT[] T[]
P[L2] D[728x90] M[320x50] OOP[F] ADUNIT[] T[]
P[R1] D[300x250] M[300x250] OOP[F] ADUNIT[] T[]
P[R2] D[300x250] M[320x50] OOP[F] ADUNIT[] T[]
DIM[1x3] LBL[article-box] SEL[] IDX[] TMPL[standalone] T[]
P[R3] D[300x250] M[0x0] OOP[F] ADUNIT[] T[]

Todd Neeley

Todd Neeley
Connect with Todd: