Using data has been an important element of Bayer Crop Science for decades, but the organization got a real data-driven shot in the arm with Bayer’s 2018 acquisition of Monsanto. Since then, it has sought to apply machine learning and artificial intelligence to every aspect of its business.
One area where AI is playing an increasingly important role is precision agriculture. Traditional farming today means planting your field with seeds, spraying the entire field with fertilizer and pesticides, and irrigating the entire field. But fields aren’t uniform: The soil isn’t necessarily the same across the field, some parts need more fertilizer or less; some areas need more water or less; and so on. Precision agriculture seeks to bring more intelligence to the process.
“For precision planting and precision farming, companies like Bayer work really closely with the farmers to understand their land, their acreage, the type of soil they have, the water flow, and then work with them,” says Michelle Lacy, data strategy lead for R&D in the Plant Biotechnology Division at Bayer Crop Science. “These are the best seeds; these are the best plant breeds that work well in this area of your farm. Planting the right seed type — the right corn line or soybean line or what have you — that will thrive in that soil type or the amount of water or nitrogen that you have. That’s precision farming.”
This data-driven approach to agriculture requires data to be collected by farmers and their equipment, such as combines that take soil measurements as they work; third parties, including academics and government employees; and Bayer Crop Science employees.
Bayer is now leveraging image analytics powered by TIBCO Spotfire visual analytics as part of that effort. It uses drones that take high-definition pictures to monitor crops. Plant height, color/greenness, and whether plants are diseased or infested with insects can be captured by images. A machine learning algorithm processes those images and blends the data with data about the soil, irrigation, and fertilization to provide in-depth insights into each part of a field. Bayer Crop Science uses TIBCO Data Virtualization software to understand the various data sources and schemas and bring them together without altering the physical source.
“In the past you’d have folks go in and walk the field and take measurements and observations on clipboards,” Lacy says. “They still do that to some extent, but for large data capture, that’s the use case where we use drones.”
Reaping value from drone data
In addition to helping farmers, that data also drives Bayer Crop Science’s research into new plant strains. The images can be used to compare an insect-resistant or disease-resistant strain against control groups. In areas affected by tornados, AI can stitch together drone images to show which crops stood up or fell over during storms.
“It’s hard to know how much damage you have just by eyeballing it because corn plants are really tall,” Lacy says. “So being able to fly over it and assess the damage is another use case for image analytics and our drones. It’s just faster. We can get information either to our labs or our farmers a lot quicker.”
Lacy says one key to Bayer Crop Science’s data-driven transformation has been its adherence to FAIR (findable, accessible, interoperable, and reusable) data principles. FAIR is a sort of data bill of rights that says users should be able to find their data easily, users should be able to access the data they need when making decisions (while still following cyber security policies), data should be interoperable, and data should be reusable.
“It’s extremely important,” Lacy says. “It’s the foundation of our data strategy.”
Often, data developed by one group can be helpful to work undertaken by other groups. To make efficient use of the data, the various groups need to know the data exists and how to find it, and the data must be compatible.
“If you’re running various assays on a single plant, whether it’s field assays or you’re running different experiments in a lab, you’ve got to be able to bring that data together,” Lacy says. “You can think of it as a jigsaw puzzle and all these different assays you run are pieces of that puzzle. The project lead has to put those pieces back together.”