Empowering the agriculture value chain

1. EMPOWERING
The Agriculture Value Chain
Claudia Roessler
Director Agriculture Industry
Microsoft Corporation

2. Sustainable production
of global greenhouse emissions come
from Agriculture24%
of global water resources are needed
for Agriculture70%
liters of water to be saved in 2030 from
implementing Smart Agriculture251T
Global access
Reduce poverty for 65% of the world’s
poor who live in rural areas and work in
farming
65%
1 in 9 people are undernourished (UN)
70% more food is needed by 2050
Economic growth
Agriculture contributes 10% of
global GDP and up to 30% in low
income countries
10-30%
of global workers are employed
by Agriculture30%
Global Agriculture revenue$4.8T
Source: Sustainability development goals, UNITED NATIONS 2017
Agriculture needs to fundamentally transform

3. Sensor ModuleFarm Mixed RealityProduct
Fueling the Agriculture Decision Engine
GeographyEquipment

4. Questions
• Monitor farm conditions remotely and to improve inputs?
• Monitor equipment to ensure proper functioning?
• Remote manage equipment?
• Make use of public data & aggregated insight?

5. Smart Irrigation – advanced
analytics-enabled maintenance
• Preserved 30% of water resources
• Increased energy savings up to 50%
• Analyzed spot prices to perform pumping
when energy is at its cheapest.

7. Sensor ModuleFarm Mixed RealityProduct
Fueling the Agriculture Decision Engine
GeographyEquipment

8. Questions
• How do I overcome connectivity & low bandwidth challenges?
• Gain insight from siloed data/isolated systems?
• Make data-driven agriculture affordable?
• Train non-technical people?

9. FarmBeats
White Space
Unlicensed, long-range,
and low-cost TV White
Space radios connect
sensors and capture data
Heat Map
Algorithms process
data to create a
detailed heat map of
the field
Local Farm Computer
Windows 10 IoT-based
Azure IoT Gateway performs
computations and AI at the
FarmBeats Edge in the
farmer’s house/office
Farmer
Heat map data is sent
to a farmer who
applies insight to
increase efficiency and
conserve resources
Soil Sensor
Solar-powered soil
sensors gather moisture,
soil chemistry, and
temperature data
TYE Balloons
Low-cost Tethered Eye
helium balloons
continuously stream
images and video of the
farm to the Azure IoT
Gateway in places where
drones are not permitted
Drones
Autopilot drones gather
aerial imagery (visible
and multi-spectral)
Azure Cloud
Machine learning algorithms
integrate sensor data with aerial
imagery and crop predictions
FarmBeats provides farmers with
access to Microsoft Cloud and AI
technologies, enabling data-driven
decisions to improve agricultural
yield, lower overall costs, and reduce
the environmental impact of
agricultural production

10. Connectivity
Less cloud dependency
Sensor/Image fusion ©images by Gates Notes
Microsoft FarmBeats
@Dancing Crow Farm
• 30% less water for irrigation
• 44% less lime to control soil pH
• Moisture level determine
planting of seeds
• Identify inadequate drainage
impacting produce quality
• Flooding control and patterns
Predictions

11. Latency requirements
11
Water puddle Cow excreta Cow Herd Stray cow

12. Leveraging machine learning & AI
https://youtu.be/yJs8LBt6n_s

14. Questions
• How can I provide context based advise to farmers?
• How can I use text, voice, and images to improve data collection?
• How can I use image data to assess broad geographical areas?

15. AI for
Agriculture
advisory

16. Giving organizations a faster, more
effective, and lower cost land cover
mapping tool to help them better analyze,
monitor, and manage natural resources.
// AI for Earth
Land cover mapping
• Automate vegetation index
• Deforestation
• Sustainable practices
Example:
recognizing and marking
visible evidence of
practices such as the
presence of filter strips
and riparian buffers.

17. Sensor ModuleFarm Mixed RealityProduct journey
Fueling the Agriculture Decision Engine
Geographical areaEquipment

18. Questions
• How can I track food from “Farm to
Table”?
• Create less food waste and loss?
• Adhere to compliance?
• Deal with food recalls?
• Ensure sustainability tracking? nge

19. Bühler is ensuring food
safety with Blockchain
• Augmented physical machines with an AI-enabled
insights platform work to ensure all grains processed are
toxin-free
• Multilayer QR code labels to track grain and ensure it’s
handled according to regulations.
• Digitally verify that all grains are handled compliantly as
they move through the supply chain, transparency to
each counterparty through a shared ledger
• Assess any contamination and alert all parties in real-
time to avoid a public safety issue.
We’re not only going to get economic results, but hopefully we can
transform and save lives at the same time. It’s the most exciting and
valuable project I’ve worked on in my 40 years with the company. It’s
a really big thing.
— Ben Deefholts
Senior research engineer, Bühler

20. Blockchain in action: real-time grain tracking
Food processor A has
breached the contract and
is liable for penalty
SHARED LEDGER
Smart contract
created
Grain needs to be
harvested and
shipped.
Granary
• Grains stored
• Compliant
handling
Miller
• Grains milled
• Compliant
handling
Food processors
A: Pasta products
B: Bread products
C: Rice products
Retail store
A: Rejected – non-
compliant handling
B: Accepted
C: Accepted
Harvest
• 25% of grains
have toxins
• 75% of grains
are shipped
Carriers
• Grains delivered
• Compliant
handling
Carrier 1
The product is securely shipped with QR codes that
verify that shipping and handling happen per
manufacturer’s regulations and compliance standards.
Miller
The grains are milled and
must be handled per
manufacturer’s regulations
and compliance standards.
Granary
The grains are stored per
manufacturer’s regulations and
compliance standards.
Retail store
Rejects the pasta products sent by food processor
A due to poor handling procedures
Harvest
The grains are harvested and assessed
for contaminants. The remaining
grains are then packaged with
multilayer QR codes.
Food processors
Transform the grains into
packaged goods for consumption
per manufacturer’s regulations
and compliance standards .
A
B
C
25% of grains are
found to have toxins
and are rejected.
Carrier 2Carrier 3

21. Sensor ModuleFarm Mixed RealityProduct journey
Fueling the Agriculture Decision Engine
Geographical areaEquipment

23. Simulation and Prediction
Scenario One: High Yields, Average ROI Scenario Two: Average Yield, High ROI

24. Thank you
Claudia Roessler https://www.linkedin.com/in/claudia-rössler-microsoft/
Director Agriculture Industry
Microsoft Corporation