big data analytics in the animal production domain - | icar...claudia kamphuis, yvette de haas, erik...
TRANSCRIPT
![Page 1: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/1.jpg)
Big Data analytics in the animal production domain
19th of June, 2019 – ICAR, Prague
Claudia Kamphuis, Erwin Mollenhorst & Roel Veerkamp
![Page 2: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/2.jpg)
Big Data
1.79 billion 317 million
monthly active users
![Page 3: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/3.jpg)
Gartner’s hype cycle
3
![Page 4: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/4.jpg)
Big Data in Animal Agriculture?
Example projects
Key pointers to make Big Data useful
Outline
4
![Page 5: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/5.jpg)
Big data field?
Volume
Velocity
Variety
Veracity
Variability
Value
5
![Page 6: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/6.jpg)
Tractors
Tillage equipment
Milking robot / parlour
Feed boxes
.....
Sources of Big Data - Machines
6
![Page 7: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/7.jpg)
Soil analysis
Soil type
Soil temperature
Ground water level
Crop history
.....
Sources of Big Data - Fields
7
![Page 8: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/8.jpg)
Genomic data
Sensors / images
● ID
● Behaviour
● Health
● Position
● Smart fencing
.....
Sources of Big Data - Animals
8
![Page 9: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/9.jpg)
Gaseous emissions
● Methane (CH4)
● Ammonium (NH3)
● Nitrous oxide (N2O)
Ground/surface water
Weather
.....
Sources of Big Data - Environment
9
![Page 10: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/10.jpg)
Slaughter data
Tracking & tracing
Farm management program
Financial accounts
.....
Sources of Big Data – production chain
10
![Page 11: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/11.jpg)
Big Data in Animal Agriculture?
Four example projects
Key pointers to make Big Data useful
Outline
11
![Page 12: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/12.jpg)
Predict lifespan of an animal still alive
combining genomics and DHI
12
![Page 13: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/13.jpg)
DNA: breeding value for 50 traits
72 additional phenotypic records; Pedigree, dam, own birth and
calving records, test milk days, movement (transport),
inseminations, viability & vitality of calves, survival status at various
points, farm...
Statistical methods: regression, naive bayes, random forest
Dairy cow’s longevity
13
![Page 14: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/14.jpg)
Better management predicting longevity
14
Combination of genomic breeding values and phenotypic traits important to predict survival, even after first calving
![Page 15: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/15.jpg)
Resilience and efficiency of animal and farms
15
![Page 16: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/16.jpg)
Resilience
16
Resilience through the theory of critical transitions Scheffer et al., 2012
Stable
state 1
Stable
state 2
Perturbation
Stable
state 1
Stable
state 2
Perturbation
![Page 17: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/17.jpg)
Breeding for resilience using daily yield data
17
![Page 18: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/18.jpg)
Environmental impact
Manure management
Erwin Mollenhorst, Claudia Kamphuis, Gerard Migchels
18
![Page 19: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/19.jpg)
Current situation:
● Fixed phosphate application norms for crops / grassland
● 3 classes, based on P status of field
● For crops: 50 / 60 / 75 kg P2O5 (app. 22 / 26 / 33 kg P)
Can we predict future maize yields (= P) based on farm data and open
source weather data using artificial intelligence?
Environmental norms
19
![Page 20: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/20.jpg)
BodemHack, May 2018, De Marke
Ideas developed at Hackatons
20
(Be)MestWijs won the incentive prize for most market-ready resultJob de Pater (NMI), Reinier Wieringa(EZ-Dictu), Erwin Mollenhorst (WUR), Justin Steenhuis (VAA ICT), Herbert Meuleman (CRV), Claudia Kamphuis and Gerard Migchels (both WUR). Not on foto: Roel Veerman (Akkerweb)
MestHack October 2017, Dairy Campus MaxiMy-N won with a data- en IT-
implementation to measure and show
ecosystem services
Mehrab Marri (MSc), Joost Lahr, Henk Janssen,
Yke van Randen, Erwin Mollenhorst (all 4 WUR)
and Lucas vd Zee (UvA). In front: Gerard Ros
(NMI) and Charon Zondervan (jury)
![Page 21: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/21.jpg)
Crop in previous year
(grass/maize)
Phosphate status field Maximum temperature
in July
Average Pyield maize
same field past 7 yrs
Most important combining data sources
21
Cropping
scheme
Soil status Weather Yield history
![Page 22: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/22.jpg)
A flexible data architecture to automate
collection of (near) real-time methane sensor
data at commercial dairy farms
Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger
Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp
![Page 23: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/23.jpg)
Collect and visualise methane data flexible and
automated on commercial dairy farms
NB-IoT
Data push
every 3 min
![Page 24: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/24.jpg)
Big Data in Animal Agriculture?
Example projects
Lessons learned and key pointers to make Big Data useful
Outline
24
![Page 25: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/25.jpg)
Making data available for
the benefit of ...
farmer
consultant
legislation
technology provider
....
1) Organise data availability across sources
25
van het Land, ICAR, 2017
![Page 26: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/26.jpg)
2) Domain knowledge is present (and leading)
26
Artificial intelligence
Machine learning
Data lakes
Cloud computing
Block chain
farming
animal health,
food production
![Page 27: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/27.jpg)
3) Other people & ways of working e.g. hackatons
27
Hackathon smart farming,
December 2017, Westfort, Nieuwegein
Big data analytics & male fertility,
November 2017, Dairy Campus
Computer Assisted Semen Analysis
Multidisciplinary teams, not tech. only!Combining data, software, hardware and design
CompetitionPressure cooker setting
![Page 28: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/28.jpg)
More and more big data will come available
Key pointers to success
● Sharing data (who organises and benefits?)
● Domain knowledge should not be forgotten
● Domain experts should adapt
Data analytics is not the silver bullet!
Summary
28
![Page 29: Big Data analytics in the animal production domain - | ICAR...Claudia Kamphuis, Yvette de Haas, Erik van den Bergh, Gerrit Seiger Erwin Mollenhorst, Dirkjan Schokker, Roel Veerkamp](https://reader033.vdocuments.us/reader033/viewer/2022051511/600b57afc01e2a56474e316c/html5/thumbnails/29.jpg)
Take home
Success in Big Data is not
about technical tools, but
connecting the tools with
people and domain experts
29@RFVeerkamp