PERAN PEMBELAJARAN MESIN PADA PERTANIAN ERA DIGITAL
Universitas Budi Luhur – 26 MEI 2018
TANTANGAN PERTANIAN MASA DEPAN
PERTANIAN ERA DIGITAL
PEMBELAJARAN MESIN
PERAN PEMBELAJARAN MESIN PADA PERTANIAN
ERA DIGITAL
Area Riset di Bidang Pertanian
Contoh Riset Pembelajaran Mesin di Bidang Pertanian
1
2
3
4
Chemicalfertilizer heavy
overgrazing
deforestation
5
Chemicalfertilizer heavy
deforestation
overgrazing
Flash-flood event
Increasing
drought
25% DEGRADED FARMLAND, 2050
6
Chemicalfertilizer heavy
deforestation
overgrazing
Flash-flood event
Increasing
drought
25% DEGRADED FARMLAND, 2050
Hundreds of millions of people are likely to be forced to migrate over just the next 3
decades due to rapidly deteriorating farmland productivity in many regions — caused
by modern chemical- and fertilizer-heavy farming practices, overgrazing, deforestation,
flash-flood events, and increasing drought
7
Agric. 1.0. early 20th century, a
labour-intensive, low productivity.
Feed a population, requires a vast
number of small farms
8
Agric. 2.0. late of
2050, Green
Revolution,
supplemental
nitrogen and
synthetic
pesticides,
fertilizer, and
specialized
machine
Agric. 1.0. early 20th century, a
labour-intensive, low productivity.
Feed a population, requires a vast
number of small farms
9
Agric. 2.0. late of
1950, Green
Revolution,
supplemental
nitrogen and
synthetic
pesticides,
fertilizer, and
specialized
machine
Agric. 3.0. precision farming,
Agric. 1.0. early 20th century, a
labour-intensive, low productivity. late 1990, Military GPS signal
available for public,
Feed a population, requires a vast
number of small farms
10
Agric. 2.0. late of
1950, Green
Revolution,
supplemental
nitrogen and
synthetic
pesticides,
fertilizer, and
specialized
machine
Agric 4.0. a new
boost in
precision
agriculture
(early 2010),
cheap sensors,
low cost microprpcessor, high
bandwidth
celluler comm.,
Agric. 3.0. precision farming,
Agric. 1.0. early 20th century, a
cloud based
labour-intensive, low productivity. late 1990, Military GPS signal
ICT
available for public,
Feed a population, requires a vast
11
number of small farms
Agriculture 3.0 (precision agriculture):
- Accuracy of the operations,
- managing in-field variations rather than treating fields as a
whole,
- give each plant exactly what it needs to grow optimally
- Optimize the agronomic output while reducing the input
- Move from cutting the costs to seeking the ways to lower
costs and enhance the quality
- Technology :
GPS-signals (1990 & 2000)
Sensing & Control (1990)
Telematics (2000)
Data management (since 1980 : farming software)
During 1961- 2004, the agriculture yields
increased 300%
12
-
The Rate of yields increased is slow
Has to produce 70% more food in 2050, compare with 2011
Using less : energy, fertilizer, pesticide
Lowering levels of GHGs
Coping with climate change
22 X
Agriculture Next Gen.
- Green one
- Science and
technology at the
heart
13
14
- a set of nature-inspired computational methodologies and
approaches to address complex real-world problems to
which mathematical or traditional modelling can be useless
for a few reasons
a. The processes might be too complex for
mathematical reasoning,
b. It might contain some uncertainties during the
process,
c. The process might simply be stochastic in nature.
d. Many real-life problems cannot be translated into
binary language (unique values of 0 and 1) for
computers to process it. Computational Intelligence
therefore provides solutions for such problems
15
16
- Understanding of the
underlying system.
- Interpretable (not a black
box)
- is intelligence
demonstrated
by machines
- a machine
mimics
"cognitive"
functions
(learning and
solving)
- Learning from data so as to be
able to automate functionality
- Not understansing, but
engineering
• STAT
• ML
• DT MNG
• AI
• CI
discovering patterns in
large data sets
involving methods at
the intersection of
machine learning,
statistics, and
database systems
Study of adaptive mechanisms to enable
intelligent in complex and changing
environment algorithm inspired by
biological process
17
CS
CI
ML
AI
DT MNG
STAT
DT SAINS
18
Domain/Populati
on/Universal Set
Model
EVALUATION
Without
taking
data
Fuzzy Inference
System, Ev.
Computation,
PSO, Ant Colony
Alg., AIS.
Training
Data
Neural Network :
MLP, SOM, LVQ,
CNN, TDNN, etc.
Fuzzy : ANFIS,
Fuzzy C-Means
Testing
data
Testing
- Feature
Extraction/Representation
- Prediction
- Exploration : finding a
useful pattern/information
- Optimization : Finding the
best object due to a
certain goals
Model
19
TYPE OF LEARNING
1. Supervised
2. Unsupervised
3. Semi Supervised
4. Reinforcement
5. Learning by Expert
CRITERIA
1. Optimization
2. Biological Inspired
GOALS
-Feature Extraction/Representation
-Prediction
-Exploration : finding a useful
pattern/information
-Optimization : Finding the best
object due to a certain goals
20
Supervised (inductive) learning
Training data includes desired outputs
Unsupervised learning
Training data does not include desired outputs
Semi-supervised learning
Training data includes a few desired outputs
Reinforcement learning
Rewards from sequence of actions
Learning by Expert
21
Computationa
l Intelligence
Algorithm
- Expected Label
- Expected Value
- etc
22
Computationa
l Intelligence
Algorithm
- Expected
Label
- Expected
Value
- Valuable
patternl
- etc
23
Domain
/Population/Unive
rsal Set
Modellig
Experts and
engineers
- Expected
object in
the stated
Domain
- Expected
value/label
24
25
CI
Algorithm
Model
OK
No
- Expected
Label
- Expected
Value
- etc
28
29
30
Biological neuron
biological neural network
Artificial neuron
Artificial Neural Network
31
Tipe 1 :
Tsukamoto
Tipe 2 :
Mamdami
Tipe 3 :
Sugeno
34
35
f1 and f2 is a
function of x and y
Using
Sugeno
36
37
Selection
Crossover
Fitness
domain
Obyek/Titik
dalam
domain
Mutation
38
Particle
Ik
X =
P =
V =
x_fitness = ?
p_fitness = ?
A particle (individual) is composed of:
Three vectors:
▪ The x-vector records the current position (location) of the particle in
the search space,
▪ The p-vector records the location of the best solution found so far
by the particle, and
▪ The v-vector contains a gradient (direction) for which particle will
travel in if undisturbed.
Two fitness values:
▪ The x-fitness records the fitness of the x-vector, and
▪ The p-fitness records the fitness of the p-vector.
39
40
41
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
50
Ant Algorithms – (P.Koumoutsakos – based on notes L. Gamberdella (www.idsia.ch)
52
AIS : models the natural immune system’s ability to
detect cells foreign the body
A computational paradigm with powerfull pattern
recognition abilities, mainly applied to anomaly detection
Pathogens
Skin
Biochem ical
barriers
Innate
im m une
response
Adaptive
im m une
response
Phagocyte
Lym phocytes
53
54
Computer Security(Forrest’94’96’98, Kephart’94, Lamont’98’01,02,
Dasgupta’99’01, Bentley’00’01,02)
Anomaly Detection (Dasgupta’96’01’02)
Fault Diagnosis (Ishida’92’93, Ishiguro’94)
Data Mining & Retrieval (Hunt’95’96, Timmis’99’01, ’02)
Pattern Recognition (Forrest’93, Gibert’94, de Castro ’02)
Adaptive Control (Bersini’91)
Job shop Scheduling (Hart’98, ’01, ’02)
Chemical Pattern Recognition (Dasgupta’99)
Robotics (Ishiguro’96’97,Singh’01)
Optimization (DeCastro’99,Endo’98, de Castro ’02)
Web Mining (Nasaroui’02,Secker’05)
Fault Tolerance (Tyrrell, ’01, ’02, Timmis ’02)
Autonomous Systems (Varela’92,Ishiguro’96)
Engineering Design Optimization (Hajela’96 ’98, Nunes’00)
55
56
IT-Based Agriculture Process
• Crop Monitoring (Satellite, Drone, GPS,
Selluler Communication)
• Precision Livestock Farming
• Nutrient Management and Budgeting
• Integrated Farming
• Integrated Pest management
• Crop Simulation
• Aerial Seeding
• Agriculture Robot
• Environmental Monitoring
• Precision Agriculture
• etc
57
- Enabling by :
cheap and improved sensors and actuator
low cost micro-processor
High bandwidth celluler communication
cloud based ICT systems
big data analytics
Integrated internal and external networking of farming
operations.
- Digital information exist for all farm sectors and processes
- Communication with external partners (suppliers and end
customers) carried out electronically
- Data transmission, processing and analysis are automated
58
22 X
- IoT : intergrated structured
and unstructured data to
into food production.
Applying machine learning
to sensor or drone data
- Automation of Skills and
workforce : remotely,
automated, risks will be
identified, issues solved
- Data-driven farming : analyzing weather, seeds, soil quality,
diseases, historical data, market trend, prices
- Chatbots: assisting farmers with answers and
recommendations on specific problems
59
Climate change:
More variable and
extreme weather
conditions
Increasing temperature
Excessive rainfall
Consequences for food
and waterborne
diseases
Consequences crop
production areas and
ecosystems
Governments:
Responsibility and concern for food supply, quality and sustainability
Regulations and policy
Management and Control of Fresh and Processed Products
Agricultural and Agroindustrial Management
Systems (Quality, Cost, Delivery, Adaptability)
Control and
Assurance
Global change:
Shift demographic,
social and economical
conditions
Towards worldwide
markets
Global sourcing of fresh
and agricultural
produce
Many movements and
long distances in supply
chains
Increasing volumes and
new harvesting area
Consumer trends:
Agroindustrial Supply Chain :
Farming
Processing
Industry
Distribution
and Retail
Agricultural products are perishable, seasonal, vary
and bulky
Risk of product contamination and deterioration
during transportation
Lot of human handling
Fresh and
Processed
Products
Increase
consumption
on fresh and
processing
products
Increase
awareness of
food safety
Change in
costumer
behavior
Change in
types of
products
Changes in
demand
Increase the
need for new
and healthy
products
By : Yandra Arkeman
Companies & Fund & InvestorExperts
Management
Gov
Farmer
personal
websites
Agriculture Eco Management
System
CRM
ERP
HRM
Farmers
Agriculture Ecosystem Portal
IoT Agriculture Solution
Agri Market
Place
CMS
Customers
Precise
Agriculture
Agri Eco
Social
network
Smart
Weather
Smart
Tools
Agriculture IoT Platform
Data Storage, Processing, Mining & Analytics
Gateways
Environmental
sensors
Farm
Weather
sensors
Weather
Soil
sensors
Tools
Water
sensors
Plant
sensors
Animal
sensors
Green
Energy
COMPUTATIONAL
INTELLIGENCE IN CLIMATE
IMPACT AND MODELLING
62
JANGKA PENDEK VARIABILITAS IKLIM
a. PRAKIRAAN PEUBAH IKLIM : CURAH
HUJAN, SUHU, KELEMBABAN,
KECEPATAN DAN ARAH ANGIN
b. DAMPAK VARIABILITAS IKLIM DI SUATU
SEKTOR
c. JENIS : PREDIKSI, KLASIFIKASI,
PEMODELAN SISTEM
DATA
SATELIT/
MODEL/RE
ANALISIS
PHENOMEN
A/IKLIM
GLOBAL
PREDIKSI IKLIM LOKAL
PREDIKSI/KLASIFIKASI
DAMPAK VARIABILITAS
IKLIM
JANGKA PANJANG CLIMATE CHANGE
a. PROYEKSI IKLIM : CURAH HUJAN, SUHU,
KELEMBABAN, KECEPATAN DAN ARAH
ANGIN MENGGUNAKAN SKENARIO
YANG DITETAPKAN OLEH IPCC
b. DAMPAK PERUBAHAN IKLIM DI SUATU
SEKTOR (PERTANIAN, KETERSEDIAAN
AIR, ENERGI, KESEHATAN, DSB
c. JENIS : PREDIKSI, KLASIFIKASI,
PEMODELAN SISTEM
DATA SATELIT/
MODEL/REANA
LISIS/OBSERV
ASI (HISTORY)
DATA GCM
SESUAI
SCENARIO
DAN MODEL
PROYEKSI
IKLIM SESUAI
SCENARIO
VISUALISASI SPASIAL
DAMPAK PERUBAHAN
IKLIM PADA SUATU SEKTOR
64
Chromosome Code
Init. population
Decode
Mutation
Evaluate
Selection
Crossover
Genetic
Operators
Fitness
Genetic Coefficients
and Their Uncertaintie
DSSAT : a Crop Modelling
Simulation System
66
Data Model
Variables :
• Radiation
• Temperature
• Evaporation
• Humidity
• Precipitation
• Wind
• etc
Temporal (Historical & Future)
Spatial (latitude x longitude)
Vertical (Above and Below the
Earth’s Surface)
Volume
&
Velocity
67
Observation
Data
Climate Index :
- El Nino Southern Oscillation (ENSO),
- Indian Ocean Dipole-Mode (IOD),
- Southern Oscillation Index (SOI)
- El Nino Modoki (EMI)
etc
Global Data
Model
Influenced to
Regional Climate
Evidence
Obser
vation
Agric.
IMPACT :
Sectors
Avia.
Health
Energy
ETC
Total Hujan
Peubah Iklim Lokal :
Awal dan akhir Musim
Hujan, Total Hujan,
Jumlah Hari Hujan,
Distribusi d hari hujan
Berturut-turut, Hujan
ekstrem, panjang musim
kering, dsb.
AWAL MUSIM
HUJAN
AKHIR MUSIM
HUJAN
Series of
Data Model
Domain
Downsclaing :
Observation=f(Data Model)
Observati
onal Data
Series
1. – . 2017. Agriculture 4.0- Ensuring Connectivity of Agriculture Equipment.
FarmNet
2. Clercq, M..D., Anshu V., dan Alvaro B. Agriculture 4.0 : The Future
Technology. World Government Summit.
www.worldgovernmentsummit.org.
3. -. Industry 4.0 : Digital Transformation in Vietnam Agriculture & Indstri
Development. SaoBacDau Tachnlogy Group.
4. -. Digital Farming : What does it Really Mean?. European Agricultural
Machinery (CEMA).
5. Edward L dan Bart G. 1996. Climate & Weather Explained. Routledge,
29 West 35th Street, New York.
6. Yi-Fan Chang. 2011. An Overview of Machine Learning. (Lecture Note).
7. Pedro D. Machine Learning. (Lecture Note).
8. Jong Y.C. Machine Learning and Statistical Analysis. (Lecture Note)
9. Neuro-Fuzzy and Soft Computing : A Computational approach to learning
and Machine Intelligence. Prentice Hall. International Edition.
10. Laurene F. 1994. Fundamentals of Neural Networks. Prentice Hall,
Englewood Cliffs
11. Zbigniew M. 1996. Genetic Algorithms + Data Structures = Evolution
Programs. Th.. Ed. Ketiga. Springer
12. Engelbrecth, A.P. 2007. Computational Intellegence : An Introduction.
John Wiley & Sons, Ltd.
72
73
TANTANGAN PERTANIAN MASA DEPAN
PERTANIAN ERA DIGITAL
PEMBELAJARAN MESIN
PERAN PEMBELAJARAN MESIN PADA PERTANIAN
ERA DIGITAL
Area Riset di Bidang Pertanian
Contoh Riset Pembelajaran Mesin di Bidang Pertanian
1
2
3
4
Chemicalfertilizer heavy
overgrazing
deforestation
5
Chemicalfertilizer heavy
deforestation
overgrazing
Flash-flood event
Increasing
drought
25% DEGRADED FARMLAND, 2050
6
Chemicalfertilizer heavy
deforestation
overgrazing
Flash-flood event
Increasing
drought
25% DEGRADED FARMLAND, 2050
Hundreds of millions of people are likely to be forced to migrate over just the next 3
decades due to rapidly deteriorating farmland productivity in many regions — caused
by modern chemical- and fertilizer-heavy farming practices, overgrazing, deforestation,
flash-flood events, and increasing drought
7
Agric. 1.0. early 20th century, a
labour-intensive, low productivity.
Feed a population, requires a vast
number of small farms
8
Agric. 2.0. late of
2050, Green
Revolution,
supplemental
nitrogen and
synthetic
pesticides,
fertilizer, and
specialized
machine
Agric. 1.0. early 20th century, a
labour-intensive, low productivity.
Feed a population, requires a vast
number of small farms
9
Agric. 2.0. late of
1950, Green
Revolution,
supplemental
nitrogen and
synthetic
pesticides,
fertilizer, and
specialized
machine
Agric. 3.0. precision farming,
Agric. 1.0. early 20th century, a
labour-intensive, low productivity. late 1990, Military GPS signal
available for public,
Feed a population, requires a vast
number of small farms
10
Agric. 2.0. late of
1950, Green
Revolution,
supplemental
nitrogen and
synthetic
pesticides,
fertilizer, and
specialized
machine
Agric 4.0. a new
boost in
precision
agriculture
(early 2010),
cheap sensors,
low cost microprpcessor, high
bandwidth
celluler comm.,
Agric. 3.0. precision farming,
Agric. 1.0. early 20th century, a
cloud based
labour-intensive, low productivity. late 1990, Military GPS signal
ICT
available for public,
Feed a population, requires a vast
11
number of small farms
Agriculture 3.0 (precision agriculture):
- Accuracy of the operations,
- managing in-field variations rather than treating fields as a
whole,
- give each plant exactly what it needs to grow optimally
- Optimize the agronomic output while reducing the input
- Move from cutting the costs to seeking the ways to lower
costs and enhance the quality
- Technology :
GPS-signals (1990 & 2000)
Sensing & Control (1990)
Telematics (2000)
Data management (since 1980 : farming software)
During 1961- 2004, the agriculture yields
increased 300%
12
-
The Rate of yields increased is slow
Has to produce 70% more food in 2050, compare with 2011
Using less : energy, fertilizer, pesticide
Lowering levels of GHGs
Coping with climate change
22 X
Agriculture Next Gen.
- Green one
- Science and
technology at the
heart
13
14
- a set of nature-inspired computational methodologies and
approaches to address complex real-world problems to
which mathematical or traditional modelling can be useless
for a few reasons
a. The processes might be too complex for
mathematical reasoning,
b. It might contain some uncertainties during the
process,
c. The process might simply be stochastic in nature.
d. Many real-life problems cannot be translated into
binary language (unique values of 0 and 1) for
computers to process it. Computational Intelligence
therefore provides solutions for such problems
15
16
- Understanding of the
underlying system.
- Interpretable (not a black
box)
- is intelligence
demonstrated
by machines
- a machine
mimics
"cognitive"
functions
(learning and
solving)
- Learning from data so as to be
able to automate functionality
- Not understansing, but
engineering
• STAT
• ML
• DT MNG
• AI
• CI
discovering patterns in
large data sets
involving methods at
the intersection of
machine learning,
statistics, and
database systems
Study of adaptive mechanisms to enable
intelligent in complex and changing
environment algorithm inspired by
biological process
17
CS
CI
ML
AI
DT MNG
STAT
DT SAINS
18
Domain/Populati
on/Universal Set
Model
EVALUATION
Without
taking
data
Fuzzy Inference
System, Ev.
Computation,
PSO, Ant Colony
Alg., AIS.
Training
Data
Neural Network :
MLP, SOM, LVQ,
CNN, TDNN, etc.
Fuzzy : ANFIS,
Fuzzy C-Means
Testing
data
Testing
- Feature
Extraction/Representation
- Prediction
- Exploration : finding a
useful pattern/information
- Optimization : Finding the
best object due to a
certain goals
Model
19
TYPE OF LEARNING
1. Supervised
2. Unsupervised
3. Semi Supervised
4. Reinforcement
5. Learning by Expert
CRITERIA
1. Optimization
2. Biological Inspired
GOALS
-Feature Extraction/Representation
-Prediction
-Exploration : finding a useful
pattern/information
-Optimization : Finding the best
object due to a certain goals
20
Supervised (inductive) learning
Training data includes desired outputs
Unsupervised learning
Training data does not include desired outputs
Semi-supervised learning
Training data includes a few desired outputs
Reinforcement learning
Rewards from sequence of actions
Learning by Expert
21
Computationa
l Intelligence
Algorithm
- Expected Label
- Expected Value
- etc
22
Computationa
l Intelligence
Algorithm
- Expected
Label
- Expected
Value
- Valuable
patternl
- etc
23
Domain
/Population/Unive
rsal Set
Modellig
Experts and
engineers
- Expected
object in
the stated
Domain
- Expected
value/label
24
25
CI
Algorithm
Model
OK
No
- Expected
Label
- Expected
Value
- etc
28
29
30
Biological neuron
biological neural network
Artificial neuron
Artificial Neural Network
31
Tipe 1 :
Tsukamoto
Tipe 2 :
Mamdami
Tipe 3 :
Sugeno
34
35
f1 and f2 is a
function of x and y
Using
Sugeno
36
37
Selection
Crossover
Fitness
domain
Obyek/Titik
dalam
domain
Mutation
38
Particle
Ik
X =
P =
V =
x_fitness = ?
p_fitness = ?
A particle (individual) is composed of:
Three vectors:
▪ The x-vector records the current position (location) of the particle in
the search space,
▪ The p-vector records the location of the best solution found so far
by the particle, and
▪ The v-vector contains a gradient (direction) for which particle will
travel in if undisturbed.
Two fitness values:
▪ The x-fitness records the fitness of the x-vector, and
▪ The p-fitness records the fitness of the p-vector.
39
40
41
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
max
y
fitness
x
search space
min
50
Ant Algorithms – (P.Koumoutsakos – based on notes L. Gamberdella (www.idsia.ch)
52
AIS : models the natural immune system’s ability to
detect cells foreign the body
A computational paradigm with powerfull pattern
recognition abilities, mainly applied to anomaly detection
Pathogens
Skin
Biochem ical
barriers
Innate
im m une
response
Adaptive
im m une
response
Phagocyte
Lym phocytes
53
54
Computer Security(Forrest’94’96’98, Kephart’94, Lamont’98’01,02,
Dasgupta’99’01, Bentley’00’01,02)
Anomaly Detection (Dasgupta’96’01’02)
Fault Diagnosis (Ishida’92’93, Ishiguro’94)
Data Mining & Retrieval (Hunt’95’96, Timmis’99’01, ’02)
Pattern Recognition (Forrest’93, Gibert’94, de Castro ’02)
Adaptive Control (Bersini’91)
Job shop Scheduling (Hart’98, ’01, ’02)
Chemical Pattern Recognition (Dasgupta’99)
Robotics (Ishiguro’96’97,Singh’01)
Optimization (DeCastro’99,Endo’98, de Castro ’02)
Web Mining (Nasaroui’02,Secker’05)
Fault Tolerance (Tyrrell, ’01, ’02, Timmis ’02)
Autonomous Systems (Varela’92,Ishiguro’96)
Engineering Design Optimization (Hajela’96 ’98, Nunes’00)
55
56
IT-Based Agriculture Process
• Crop Monitoring (Satellite, Drone, GPS,
Selluler Communication)
• Precision Livestock Farming
• Nutrient Management and Budgeting
• Integrated Farming
• Integrated Pest management
• Crop Simulation
• Aerial Seeding
• Agriculture Robot
• Environmental Monitoring
• Precision Agriculture
• etc
57
- Enabling by :
cheap and improved sensors and actuator
low cost micro-processor
High bandwidth celluler communication
cloud based ICT systems
big data analytics
Integrated internal and external networking of farming
operations.
- Digital information exist for all farm sectors and processes
- Communication with external partners (suppliers and end
customers) carried out electronically
- Data transmission, processing and analysis are automated
58
22 X
- IoT : intergrated structured
and unstructured data to
into food production.
Applying machine learning
to sensor or drone data
- Automation of Skills and
workforce : remotely,
automated, risks will be
identified, issues solved
- Data-driven farming : analyzing weather, seeds, soil quality,
diseases, historical data, market trend, prices
- Chatbots: assisting farmers with answers and
recommendations on specific problems
59
Climate change:
More variable and
extreme weather
conditions
Increasing temperature
Excessive rainfall
Consequences for food
and waterborne
diseases
Consequences crop
production areas and
ecosystems
Governments:
Responsibility and concern for food supply, quality and sustainability
Regulations and policy
Management and Control of Fresh and Processed Products
Agricultural and Agroindustrial Management
Systems (Quality, Cost, Delivery, Adaptability)
Control and
Assurance
Global change:
Shift demographic,
social and economical
conditions
Towards worldwide
markets
Global sourcing of fresh
and agricultural
produce
Many movements and
long distances in supply
chains
Increasing volumes and
new harvesting area
Consumer trends:
Agroindustrial Supply Chain :
Farming
Processing
Industry
Distribution
and Retail
Agricultural products are perishable, seasonal, vary
and bulky
Risk of product contamination and deterioration
during transportation
Lot of human handling
Fresh and
Processed
Products
Increase
consumption
on fresh and
processing
products
Increase
awareness of
food safety
Change in
costumer
behavior
Change in
types of
products
Changes in
demand
Increase the
need for new
and healthy
products
By : Yandra Arkeman
Companies & Fund & InvestorExperts
Management
Gov
Farmer
personal
websites
Agriculture Eco Management
System
CRM
ERP
HRM
Farmers
Agriculture Ecosystem Portal
IoT Agriculture Solution
Agri Market
Place
CMS
Customers
Precise
Agriculture
Agri Eco
Social
network
Smart
Weather
Smart
Tools
Agriculture IoT Platform
Data Storage, Processing, Mining & Analytics
Gateways
Environmental
sensors
Farm
Weather
sensors
Weather
Soil
sensors
Tools
Water
sensors
Plant
sensors
Animal
sensors
Green
Energy
COMPUTATIONAL
INTELLIGENCE IN CLIMATE
IMPACT AND MODELLING
62
JANGKA PENDEK VARIABILITAS IKLIM
a. PRAKIRAAN PEUBAH IKLIM : CURAH
HUJAN, SUHU, KELEMBABAN,
KECEPATAN DAN ARAH ANGIN
b. DAMPAK VARIABILITAS IKLIM DI SUATU
SEKTOR
c. JENIS : PREDIKSI, KLASIFIKASI,
PEMODELAN SISTEM
DATA
SATELIT/
MODEL/RE
ANALISIS
PHENOMEN
A/IKLIM
GLOBAL
PREDIKSI IKLIM LOKAL
PREDIKSI/KLASIFIKASI
DAMPAK VARIABILITAS
IKLIM
JANGKA PANJANG CLIMATE CHANGE
a. PROYEKSI IKLIM : CURAH HUJAN, SUHU,
KELEMBABAN, KECEPATAN DAN ARAH
ANGIN MENGGUNAKAN SKENARIO
YANG DITETAPKAN OLEH IPCC
b. DAMPAK PERUBAHAN IKLIM DI SUATU
SEKTOR (PERTANIAN, KETERSEDIAAN
AIR, ENERGI, KESEHATAN, DSB
c. JENIS : PREDIKSI, KLASIFIKASI,
PEMODELAN SISTEM
DATA SATELIT/
MODEL/REANA
LISIS/OBSERV
ASI (HISTORY)
DATA GCM
SESUAI
SCENARIO
DAN MODEL
PROYEKSI
IKLIM SESUAI
SCENARIO
VISUALISASI SPASIAL
DAMPAK PERUBAHAN
IKLIM PADA SUATU SEKTOR
64
Chromosome Code
Init. population
Decode
Mutation
Evaluate
Selection
Crossover
Genetic
Operators
Fitness
Genetic Coefficients
and Their Uncertaintie
DSSAT : a Crop Modelling
Simulation System
66
Data Model
Variables :
• Radiation
• Temperature
• Evaporation
• Humidity
• Precipitation
• Wind
• etc
Temporal (Historical & Future)
Spatial (latitude x longitude)
Vertical (Above and Below the
Earth’s Surface)
Volume
&
Velocity
67
Observation
Data
Climate Index :
- El Nino Southern Oscillation (ENSO),
- Indian Ocean Dipole-Mode (IOD),
- Southern Oscillation Index (SOI)
- El Nino Modoki (EMI)
etc
Global Data
Model
Influenced to
Regional Climate
Evidence
Obser
vation
Agric.
IMPACT :
Sectors
Avia.
Health
Energy
ETC
Total Hujan
Peubah Iklim Lokal :
Awal dan akhir Musim
Hujan, Total Hujan,
Jumlah Hari Hujan,
Distribusi d hari hujan
Berturut-turut, Hujan
ekstrem, panjang musim
kering, dsb.
AWAL MUSIM
HUJAN
AKHIR MUSIM
HUJAN
Series of
Data Model
Domain
Downsclaing :
Observation=f(Data Model)
Observati
onal Data
Series
1. – . 2017. Agriculture 4.0- Ensuring Connectivity of Agriculture Equipment.
FarmNet
2. Clercq, M..D., Anshu V., dan Alvaro B. Agriculture 4.0 : The Future
Technology. World Government Summit.
www.worldgovernmentsummit.org.
3. -. Industry 4.0 : Digital Transformation in Vietnam Agriculture & Indstri
Development. SaoBacDau Tachnlogy Group.
4. -. Digital Farming : What does it Really Mean?. European Agricultural
Machinery (CEMA).
5. Edward L dan Bart G. 1996. Climate & Weather Explained. Routledge,
29 West 35th Street, New York.
6. Yi-Fan Chang. 2011. An Overview of Machine Learning. (Lecture Note).
7. Pedro D. Machine Learning. (Lecture Note).
8. Jong Y.C. Machine Learning and Statistical Analysis. (Lecture Note)
9. Neuro-Fuzzy and Soft Computing : A Computational approach to learning
and Machine Intelligence. Prentice Hall. International Edition.
10. Laurene F. 1994. Fundamentals of Neural Networks. Prentice Hall,
Englewood Cliffs
11. Zbigniew M. 1996. Genetic Algorithms + Data Structures = Evolution
Programs. Th.. Ed. Ketiga. Springer
12. Engelbrecth, A.P. 2007. Computational Intellegence : An Introduction.
John Wiley & Sons, Ltd.
72
73