Experts predict that the human population is likely to reach an estimated 8.6 billion in 2030 and 9.8 billion in 2050[i]. How will farmers of tomorrow keep up? The current rate of food production is simply not enough to feed our expanding population. John Deere (“the Company”) has been working hard to solve this agricultural problem by introducing new technology and maximizing efficiency across the agricultural supply chain. The mission to increase agricultural yield has also pushed the Company to innovate and leverage Machine Learning.
Within its Agricultural and Turf business segment, John Deere has been pursuing a more focused strategy around precision agriculture. Precision agriculture employs a variety of embedded and connected technologies that rely on remote sensing, global positioning systems, and communication systems to generate big data and apply machine learning and predictive analytics. These technologies and insights allow for more precise application of agricultural management inputs such as fertilizer, seeds, and pesticides, resulting in lower costs and improved yields[ii]. For example, using sensor-driven measures or vital statistics, machine learning algorithms can estimate the well-being of a plant and determine the amount of fertilizer to apply to a specific patch of land to maximize yield[iii]. Machine learning can also be used for plant water stress detection or for soil erosion assessment.
Doubling down on its commitment to precision agriculture, in September 2017, John Deere made a key technology acquisition for $300 million dollars in the field of machine learning: Blue River Technology. This acquisition not only allowed John Deere to significantly increase its technical competency, but it also strengthened its competitive edge.
Blue River Technology, an early staged company, developed See and Spray, which empowers growers to make decisions using data at an individual plant level. Its technology leverages computer vision and machine learning algorithms to detect, identify, and make management decision about every plant in the field. See and Spring can precisely identify herbicides and spray chemicals, which gives farmers an unprecedented level of visibility and control, and it can even help them prevent herbicide-resistant weeds while eliminating 90% of the herbicide volumes that growers spray[iv]. This technology not only helps reduce chemical wastes but also reduces the cost of production that will be passed down to the end consumers.
Adopting new technology is hard. Trying to educate farmers who are not technically savvy and change behavior can be even more difficult. In the near term, John Deere should focus on proving out the value of these technologies and work on simplifying and operationalizing their use in the field with the farmers. Also, John Deere should engage farmers not just in the beginning of the production cycle with an out-of-the-box do-it-yourself solution, but throughout the harvest season so that they can more consistently contribute to the big data pool and collectively benefit from the insight generated.
In the longer term, John Deere should deepen its technical connection with its farmers to collect, control, and manage even more data that can be generated in other areas to further increase yield and decrease costs. John Deere should be investing heavily in R&D to develop more smart sensors that can be integrated into all of its tractors, equipment, or be installed across the land[v]. It should strive for cloud connectivity for all of its devices and create a platform that others can build on or contribute to, including integration with other sources of data such as water supply, weather, climate change, and plant diseases. John Deere can position itself to lead the next agricultural transformation through machine learning and even artificial intelligence to help build smart farms that will meet the demand of future generations.
In the context of John Deere and its product innovation in precision agriculture, two questions come to mind:
- How can the Company help train a new breed of farmers? It takes smart users to bring the maximum value out of great technology. How can the Company ensure that its customers are ready to join this movement and do not simply consume but also contribute to more advancement? How can the Company build a community of innovators?
- Farmer operations are fundamentally rural, yet precision agricultural and machine learning technology is built on distributed sensor networks that require a high level of data transfer and connectively. How can John Deere scale the technology without reliable rural broadband network or even lack of connectivity in the rural area?
Click on this link to see See & Spray – Blue River Technology’s precision weed control machine
[i] John Deere. (2017). John Deere Strategy presentation December 2017. Moline, IL: John Deere Corporate Presentation.
[ii] (DHS), O. o. (2018). Threats to Precision Agriculture. 2018 Public-Private Analytic Exchange Program. Deere and Company. (2017). 2017 John Deere Annual Report. Deere and Company.
[iii] Robotics: State of the Art and Future Challenges. Artificial Intelligence, Volume 172, Issue 18, December 2008, Pages 1967-1972. Christopher Stanton, Mary-Anne Williams
[iv] Technology, B. R. (n.d.). Company Overview. Introducing See & Spray: precisely spraying herbicides, p. http://www.bluerivertechnology.com.
[v] Precision Agriculture Technology and Robotics for Good Agricultural Practices. IFAC Proceedings Volumes, Volume 46, Issue 4, 2013, Pages 1-4. Josse De Baerdemaeker