Artificial Intelligence and Heuristics for Enhanced Food Security

Preface
Acknowledgments
Contents
Abbreviations
Part I: Introduction to Artificial Intelligence and Heuristics
	Chapter 1: Introduction
		1.1 Exploratory Data Analysis (EDA)
			1.1.1 Vegetation Health Index: Exploratory Data Analysis
			1.1.2 US Vegetation Health Index Analysis
				1.1.2.1 Step 1: Load Required Libraries
				1.1.2.2 Step 2: Load Vegetation Health Index Data
				1.1.2.3 Step 3: Observe Vegetation Index Distribution
				1.1.2.4 Step 4: VHI by Province
		1.2 Machine Learning Technique: Association Mining
			1.2.1 Support
			1.2.2 Confidence
			1.2.3 Frequent Itemset
		1.3 Risk Modeling: Wheat Import Origins: Food Security
			1.3.1 New Import Origins Due to Tariffs
			1.3.2 Step 1: Data Mine Historical Import Transaction Records
			1.3.3 Step 2: Construct Data Frame to Perform Pattern Analysis
			1.3.4 Step 3: Select Wheat Commodity as Part of the Data Frame
			1.3.5 Step 4: Perform EDA (Exploratory Data Analysis)
			1.3.6 Step 5: Pivot the Table to Get Transaction Record for the Association Analysis
			1.3.7 Step 6: Save the Import Origin Records
		1.4 Risk Modeling: Association Rules to Overcome Food Insecurity
			1.4.1 Step 1: Load Egypt Import Origins 1986-2020
			1.4.2 Step 2: Perform Association Rules
			1.4.3 Step 3: Perform Above-Minimum Support and Pruning Data
			1.4.4 Step 4: Run the Association Rules
		1.5 Machine Learning Technique: Clustering
			1.5.1 Types of Data in Cluster
				1.5.1.1 Data Matrix (Or Object-by-variable Structure)
				1.5.1.2 Dissimilarity Matrix (Or Object-by-object Structure)
			1.5.2 Distance Calculations
			1.5.3 Clustering Approaches
				1.5.3.1 Hierarchical Clustering: Merging Clusters
				1.5.3.2 Non-hierarchical Clustering
		1.6 Insecurity in the World´s Food Supply: New Import Countries Based on Cluster Technique
			1.6.1 Step 1: Wheat-Producing Countries
			1.6.2 Step 2: Top Exporting Countries
			1.6.3 Step 3: Available Capacity to Export
		1.7 Machine Learning Cluster Model: Wheat Import New Origins
			1.7.1 Step 1: Load Worldwide Wheat Production (Million Tonnes) and Export (Million Tonnes) Data
			1.7.2 Step 2: Construct Dendrogram
			1.7.3 Step 3: Construct Hierarchical Cluster
			1.7.4 Step 4: What-If Modeling
		1.8 Machine Learning Insights
			1.8.1 Descriptive Analytics: Satellite Vegetation Index
			1.8.2 Predictive Analytics: Food Commodities and Food Insecurity Linkage
			1.8.3 Prescriptive Analytics: Economic Cycles and Gold Price Linkage
			1.8.4 Prognostics Analytics: Food Security and Vegetation Stress Indicators
		References
	Chapter 2: Heuristics
		2.1 Heuristics
			2.1.1 Model Formulation
			2.1.2 Solvers
				2.1.2.1 Pyomo
				2.1.2.2 GLPK Solver
				2.1.2.3 CBC Solver
				2.1.2.4 Z3 Solver
				2.1.2.5 Google Operation Research (OR) Tools
			2.1.3 Decision Variables or Variables
			2.1.4 Constraints
			2.1.5 Objective Function
		2.2 Use Case: Surgical Sutures
		2.3 Solvers and Comparison
			2.3.1 Variable Creation
			2.3.2 Constraint Creation
			2.3.3 Solve
		2.4 Linear Optimization or Linear Programming
			2.4.1 Crop Land: Profit Maximization and Optimization Using Linear Programming
			2.4.2 Solver Code: Farm Profit Maximization
			2.4.3 Solving Graphically
			2.4.4 Predicting the Number and Quantity of Fertilizers Using Linear Programming
			2.4.5 Step 1: Create Variables
			2.4.6 Step 2: Constraints
			2.4.7 Step 3: Objective Function
			2.4.8 Step 4: Solve
		2.5 Mixed-Integer Programming (MIPs)
			2.5.1 Simple MIP Algebraic Line Problem
				2.5.1.1 Declare the MIP solver
				2.5.1.2 Define the Variables
				2.5.1.3 Define the Constraints
				2.5.1.4 Define the Objective
				2.5.1.5 Call the Solver
			2.5.2 Difference Between MIP and Linear Solving
			2.5.3 Predicting the Number and Quantity of Fertilizers Using MIP Programming
			2.5.4 Code
		2.6 Constraint Optimization or Constraint Programming (CP)
			2.6.1 Maximize Line Equation: x + 2y + 2z
				2.6.1.1 Import the Libraries
				2.6.1.2 Declare the Model
				2.6.1.3 Create the Variables
				2.6.1.4 Define the Constraints
				2.6.1.5 Define the Objective Function
				2.6.1.6 Call the Solver
			2.6.2 Risk Reduction Knapsack
			2.6.3 Complete Code
		2.7 Model and Simulation (M&S) Process and Industrialization Heuristics
		2.8 Choosing Solver Technology
		References
	Chapter 3: Data Engineering Techniques for Machine Learning and Heuristics
		3.1 Food Security Data
		3.2 Heuristics Data and Frequencies
		3.3 Data and Food Security Long Tail
		3.4 Data Enrichment
			3.4.1 Mean/Median Strategy
				3.4.1.1 Wheat Futures
			3.4.2 Simple Imputation Strategy
			3.4.3 Most Frequent or (Zero/Constant) Strategy
			3.4.4 k-NN Strategy
			3.4.5 Multivariate Imputation by Chained Equation (MICE) Strategy
		3.5 Food Security Supplier Risk Modeling: Food-Fuel Conundrum
			3.5.1 Saudi Arabia Rice Imports and Food Security
			3.5.2 National Policy on Biofuels and Food-Fuel Conundrum
				3.5.2.1 Milled Rice
				3.5.2.2 Rice Exports in 2020
				3.5.2.3 Rice Exports in 2010
				3.5.2.4 Rice Exports in 2000
				3.5.2.5 Rice Exports in 1990
		3.6 Machine Learning Model: Saudi Arabia Rice Import Origin Cluster
			3.6.1 Step 1: Rice-Producing Countries
			3.6.2 Step 2: Top Exporting Countries
			3.6.3 Step 3: Available Capacity to Export
			3.6.4 Step 4: Export Capacity After Meeting Local Government Mandates and Food Export Polices
		3.7 Machine Learning Cluster Model: Food-Fuel Conundrum
			3.7.1 Step 1: Load Worldwide Rice Production (Million Tonnes) and Export (Million Tonnes) Data
			3.7.2 Step 2: Construct Dendrogram
			3.7.3 Step 3: Construct Hierarchical Cluster
			3.7.4 Step 4: What-If Modeling
		References
Part II: Food Security Machine Learning and Heuristics Models
	Chapter 4: Food Security
		4.1 The Human Problem of Hunger
			4.1.1 Linkages Among Agriculture (A), Food (F), and Nutrition (N)
			4.1.2 Man-Made Conflicts
			4.1.3 Food Loss
			4.1.4 Economic Access
			4.1.5 Food Insecurity Experience Scale (FIES)
		4.2 Key Drivers of Food Insecurity
			4.2.1 Food Price Instability
			4.2.2 High Correlation Between Agricultural Commodities and Oil Prices
			4.2.3 Food Security and Grain Stock Levels
			4.2.4 Stronger Demand and Slow Growth in Agricultural Productivity
			4.2.5 Exchange Rate Volatility
			4.2.6 Speculations and Commodity Futures
		4.3 Food Security Indicators and Drivers
		4.4 The Global Food Security Index (GFSI) Framework
			4.4.1 Change Is Average Food Costs
		4.5 Economic Access: Food Security Predictive Model
			4.5.1 Step 1: Load Dataset
			4.5.2 Step 2: Inspect Data from a Statistical Point of View
			4.5.3 Step 3: Inspect Food Security Data Types
			4.5.4 Step 4
			4.5.5 Step 5: Economy Typology
			4.5.6 Step 6: Poverty Rate-Class Variable
			4.5.7 Step 7: Correlation Graphs
			4.5.8 Step 8: Correlation Graphs (Numerically) Computed
			4.5.9 Step 9: Categorical Variables and Plots
			4.5.10 Step 10: Split Data
			4.5.11 Step 11: Machine Learning Regressive Model
			4.5.12 Step 12: Feature Importance
			4.5.13 Step 13: XGB Regressor
			4.5.14 Agricultural Import Tariff
			4.5.15 Market Access and Agricultural Financial Services
		4.6 Machine Learning Model and Linear Programming: Fertilizer Price Prediction Using Commodity (Rice, Sorghum, Maize, and Whea...
			4.6.1 Data Sources
			4.6.2 Step 1: Load Required Libraries
			4.6.3 Step 2: Statistical Properties of Data
			4.6.4 Perform Linear Regression
			4.6.5 What-If Analysis
				4.6.5.1 Demand Spike for Crude Oil
				4.6.5.2 Demand Goes Down
		4.7 California Health and Human Services (CHHS) Food Affordability Predictive Model
			4.7.1 Dataset: California Health and Human Services Food Affordability
			4.7.2 Food Insecurity by Household Characteristics [38]
			4.7.3 Step 1: Load Required Libraries
			4.7.4 Step 2: Load Food Affordability Data
			4.7.5 Step 3: Inspect Data Frame for Column and Types
			4.7.6 Step 4: Clean Columns (Drop Columns)
			4.7.7 Step 5: Density Function for Numerical Columns
			4.7.8 Step 6: Check for Null Rows
			4.7.9 Step 7: Correlation of Affordability Ratio with Feature Parameters
			4.7.10 Step 8: Scalers
			4.7.11 Step 8: Split Data
			4.7.12 Step 9: Random Forest Regressor Model
			4.7.13 Step 10: Feature Importance
			4.7.14 Step 11: Other Regressor Performance Table
		4.8 Food Security and Technological Innovations
		4.9 Small Farm Sustainability
		4.10 Data
		References
	Chapter 5: Food Security: Quality and Safety Drivers
		5.1 Food Security: Quality and Safety Metric
			5.1.1 Dietary Diversity
				5.1.1.1 Household Dietary Diversity Score (HDDS)
				5.1.1.2 Infant and Young Child Dietary Diversity Score (IYCDDS)
				5.1.1.3 Women´s and Individual Dietary Diversity Score (WDDS and IDDS)
				5.1.1.4 Food Consumption Score (FCS)
			5.1.2 Rates of Food Insecurity vs. the Poverty Rate
		5.2 Signal Mining: Cereal Productions vs. Food Prices vs. Food Insecurity [34, 35]
		5.3 National Surveys of Dietary Intake and Nutritional Status
			5.3.1 Composition of Foods Raw, Processed, Prepared USDA National Nutrient Database for Standard Reference, Release 27
			5.3.2 Data Dictionary
				5.3.2.1 Step 1: Load Data
				5.3.2.2 Step 2: Statistical Properties of Different Food Items
				5.3.2.3 Step 3: Nutrient Analysis: Food Items with Energy K Calories Above the Mean and Protein (g) with Max Values
				5.3.2.4 Step 4: Correlation Data
			5.3.3 Exploratory Data Analysis: USDA National Nutrient Data
				5.3.3.1 Step 1: Load Libraries
					Step 2: Helper Functions to Plot per Column Distribution
				5.3.3.2 Step 3: Load USDA Nutrition Data
				5.3.3.3 Step 4: Plot per Column Distribution
				5.3.3.4 Step 5: Plot Correlation Matrix
		5.4 Household Dietary Diversity Score (HDDS): Venezuela
			5.4.1 Dataset
			5.4.2 Survey #1: Daily Food Consumption
			5.4.3 Survey #2: Weekly Food Consumption
			5.4.4 Machine Learning Model: Household Dietary Diversity Score (HDDS)-Venezuela
				5.4.4.1 Step 1: Load Data
				5.4.4.2 Step 2: Calculate FCS Score
				5.4.4.3 Step 3: FCS Score
				5.4.4.4 Step 4: Visualize Food Security Score Distribution Histogram
				5.4.4.5 Step 5: Value Counts of Categorical Features
				5.4.4.6 Step 6: Correlation
				5.4.4.7 Step 7: Prepare Data Frame for Decision Tree
				5.4.4.8 Step 8: Construct Decision Tree
				5.4.4.9 Step 9: Construct OLS Regression Model
				5.4.4.10 Step 10: Excel Model
				5.4.4.11 Step 11: Logistic Regression
				5.4.4.12 Step 12: Predict
				5.4.4.13 Step 13: Evaluate the Model
				5.4.4.14 Step 14: Classification Report
				5.4.4.15 Step 15: Develop Confusion Matrix
				5.4.4.16 Step 16: Construct Ensemble model
				5.4.4.17 Step 17: Train the Model
				5.4.4.18 Step 18: Train and Predict
				5.4.4.19 Step 19: Construct Gradient Boost Regressor and Predict the Model
		5.5 National Quality Forum: Food Insecurity Measures
			5.5.1 Social Determinant of Health (SDoH) and Population Criteria
			5.5.2 Food Security Measure Stratification
		5.6 National Health and Nutrition Examination Survey (NHANES)
			5.6.1 Family Questionnaire
			5.6.2 NHANES Food Security
		5.7 NHANES 2017-2018 Questionnaire Data Food Security Data
			5.7.1 Dataset
			5.7.2 Search NHANES Dataset
		5.8 Use Case: National Center for Health Statistics (NCHS) Data Brief No. 303-Prevalence of Depression Among Adults Aged 20 an...
			5.8.1 Step 1: Demographics Dataset-Percentage of Persons Aged 20 and over with Depression, by Age and Sex: United States, 2013...
			5.8.2 Step 2: Mental Health-Depression Screener (DPQ)
			5.8.3 Step 3: Merge DEMO and DPQ Data Frame
			5.8.4 Step 4: Create Depression Score (Score Will Be Missing If Any of the Items Are Missing)
			5.8.5 Step 5: Categorize Age
			5.8.6 Step 6: Calculate MFC Weight for 4 Years
			5.8.7 Step 7: Adults Aged 20 and over with a Valid Depression Score
			5.8.8 Step 8: Depression Score Histogram Plot
			5.8.9 Step 9: Correlation Between Key Features and Depression Score
			5.8.10 Step 10: Regression Model
			5.8.11 Step 11: Evaluate the Model
			5.8.12 Step 12: OLS Method
		5.9 Use Case: NHANES Body Measure Associations Between Body Weight and the Health and Nutritional Status (BPXSY2 vs. BPXSY1)
			5.9.1 Step 1: Load P_BMX.XPT, BPX_J.XPT, and P_DEMO.XPT
			5.9.2 Step 2: Inspect Columns of the Dataset
			5.9.3 Step 3: Filter the Data Based on Age, Select Adult, and Scatter Plot Each Important Blood Pressure Items
			5.9.4 Step 4: Scatter Plot Blood Pressure (First Reading) mm Hg and Blood Pressure (Second Reading)
			5.9.5 Step 5: Scatter Plot Blood Pressure (First Reading) mm Hg and Gender
			5.9.6 Step 6: Scatter Plot Race and Blood Pressure (First Reading) mm Hg
			5.9.7 Step 7: Correlation
			5.9.8 Step 8: Scatter Plot Systolic and Pulse Rate
		5.10 Use Case: NHANES 2019-2020 Questionnaire Instruments FSQ-Prevalence of Food Security
			5.10.1 Step 1: Load DEMO_J.XPT Dataset
			5.10.2 Step 2: Load FSQ_J.XPT Dataset
			5.10.3 Step 3: Merge Demographic and Food Security Data Frames
		5.11 Use Case: Predict Systolic Blood Pressure for Adult Men
			5.11.1 Step 1: Load NHANES Demographic Variables and Sample Weights (P_DEMO) data
			5.11.2 Step 2: Combine Data Frames Based on the Respondent Sequence Number (Drop Nulls)
			5.11.3 Step 3: Select Adult and Male Respondents
			5.11.4 Step 4: Plot BMI Scatter Plot
			5.11.5 Step 5: Correlation Matrix
			5.11.6 Step 6: Predict Systolic Blood Pressure Model
			5.11.7 Micronutrient Availability
			5.11.8 The Diet Problem
		5.12 An Open-Source Dataset on Dietary Behaviors
		5.13 Dietary Approaches to Stop Hypertension (DASH)
		References
	Chapter 6: ML Models: Food Security and Climate Change
		6.1 Agriculture and Climate Change
			6.1.1 FAO in Emergencies Countries
			6.1.2 Climate Change: Countries Most Affected in 2019
			6.1.3 Countries Most Affected in the Period 2000-2019
			6.1.4 Colombia
				6.1.4.1 Multiple Threats to Food Security
			6.1.5 Malawi
			6.1.6 Myanmar
			6.1.7 Recent Flooding Events in Myanmar
			6.1.8 Peru
			6.1.9 Pakistan
			6.1.10 Climate Impacts on Agriculture and Food Supply
			6.1.11 CMIP6 Climate Projections
				6.1.11.1 Projection Models
			6.1.12 Climate Model: Temperature Change (RCP 6.0) (2006-2100)
		6.2 Rice
		6.3 Mathematical Modeling
		6.4 Machine Learning Model: Thailand Rice Paddy Yields and Climate Change
			6.4.1 Data Sources
			6.4.2 EDA Framework
			6.4.3 Step 1: Import Libraries
			6.4.4 Step 2: Load Thailand FAO Dataset 1991-2020
			6.4.5 Step 3: Prepare Data Frame with Gross Production Value
			6.4.6 Step 4: Get Thailand Rice Paddy Yield Data
			6.4.7 Step 5: Merge Data [Thailand Production and Yield Data]
			6.4.8 Step 6: Load Thailand Weather Data
			6.4.9 Step 7: Create Combined Weather Data Frame
			6.4.10 Step 8: Apply Imputation Strategies to Fill NAN Values (Forward Fill)
			6.4.11 Step 9: Prepare Final Rice Paddy Data Frame with Weather Details for Regression
			6.4.12 Step 10: Prepare the Final Rice Paddy Data Frame with Weather Details for Regression
			6.4.13 Step 11: Analyze the Data Distribution Label and Features
			6.4.14 Step 12: Correlation Between Rice Production and Model Feature
			6.4.15 Step 13: Rice Main Season (August to October) Model
			6.4.16 Step 14: Model Explainability
			6.4.17 Step 14: CMIP6 Projections Data and SSP Scenarios
		6.5 Exposure: Temperature, Floods, and Sea-Level Rise
		6.6 Coffee
		6.7 Machine Learning Model: Vietnam Coffee Yields and Climate Change
			6.7.1 Key Climate Change Effects
			6.7.2 Coffee Agronomy
			6.7.3 Time Series for Terra MODIS-8-Day Vegetation Index
			6.7.4 Data Sources
				6.7.4.1 Temperature, Precipitation, and Weather Data
			6.7.5 Collected Temperatures for Gia Lai
			6.7.6 Step 1: Load Coffee Dataset
			6.7.7 Step 2: Inspect Production and Yield Data
			6.7.8 Step 3: Load Weather Data-Gia Lia (Precipitation, Max Temperatures, Min Temperatures, and Average Temperature)
			6.7.9 Step 4: Correlation (Weather Data and Yield/Production)
			6.7.10 Step 4: Coffee Yield Distribution
			6.7.11 Step 5: Feature Distributions-Max. Temperatures
			6.7.12 Step 5: Feature Distributions-Min. Temperatures
			6.7.13 Step 6: Feature Distributions-Precipitations
			6.7.14 Step 7: Feature Correlations-Label Column-Coffee Production (Tons)/Coffee Yield (hg/ha) vs. Climate Variables
			6.7.15 Step 8: Linear Regression Model
			6.7.16 Step 9: Pickle the Model to Call Climate Change Values
			6.7.17 Step 10: Model Explainability
			6.7.18 Step 11: CMIP6 Projections Data and SSP Scenarios
		References
Part III: Linkage Models
	Chapter 7: Food Security and Advanced Imaging Radiometer ML Models
		7.1 Satellite Radiometer
			7.1.1 Archive: Advanced Very High-Resolution Radiometer
			7.1.2 Visible Infrared Imaging Radiometer Suite (VIIRS)
			7.1.3 ECOSTRESS
			7.1.4 Defense Meteorological Satellite Program (DMSP)
		7.2 Dairy, Food Security, and Satellite Data
			7.2.1 Global Vegetation: Cropland and Vegetation Index
			7.2.2 The Normalized Difference Vegetation Index (NDVI)
			7.2.3 Greenhouse Gas Emissions
				7.2.3.1 Relationship Between GDP and Greenhouse Gas Emissions
			7.2.4 Impact of Climate on Staple Cereals-and Crop Yields
				7.2.4.1 Wheat
				7.2.4.2 Corn or Maize
				7.2.4.3 Soybeans
				7.2.4.4 Climate Models vs. Weather Prediction Models
		7.3 Mozambique
			7.3.1 Food Prices Increased Driven by Depreciation of National Currency
			7.3.2 Near-Average Cereal Import Requirements in 2021-2022
			7.3.3 Climate Change
				7.3.3.1 NOAA: Advanced Very High-Resolution Radiometer (AVHRR) and Climate Change Parameters
				7.3.3.2 Province Averaged VH
				7.3.3.3 No Noise (Smoothed) Normalized Difference Vegetation Index (SMN)
		7.4 Machine Learning Model: Mozambique Cashew Nuts Production Model
			7.4.1 Data Sources
			7.4.2 Step 1: Import Libraries
			7.4.3 Step 2: Load Mozambique Cashew Nuts FAO Dataset 1991-2020
			7.4.4 Step 3: Prepare Data Frame with Weather Data (Average Temperature, Min/Max Temperatures, and Precipitation)
			7.4.5 Step 4: Concat Weather Data into a Frame to Prepare the Data for Modeling
			7.4.6 Step 5: Merge Production and Weather Data Frames to Prepare the Data for Modeling
			7.4.7 Step 6: Merge Data-The Final Data Frame
			7.4.8 Step 7: Correlation Values
			7.4.9 Step 8: Correlation Matrix
			7.4.10 Step 9: Prepare the Final Cashew Nut Data Frame with Weather Details for Regression
			7.4.11 Step 10: Regression Metrics
			7.4.12 Step 11: Model Explainability
		7.5 Machine Learning Model: Mozambique Cashew Nuts and Climate Projections Model
			7.5.1 CMIP6 Projections Data and SSP Scenarios
			7.5.2 Climate Model: Temperature Change (RCP 6.0) (2006-2100)
				7.5.2.1 SSP5-8.5
				7.5.2.2 SSP1-1.9
				7.5.2.3 SSP3-7.0
		7.6 Machine Learning Model: Mozambique Cashew Nuts and Normalized Difference Vegetation Index (NDVI) Model
			7.6.1 NDVI Anomaly
			7.6.2 Step 1: Load Mozambique Vegetation Health Index (VHI) DEKAD Data
			7.6.3 Step 2: Prepare Production Data Frame
			7.6.4 Step 3: Merge Cashew Nuts and MODIS VHI
			7.6.5 Step 4: Merged Data Frame Correlation Matrix
			7.6.6 Step 5: Prepare Data for Regression Model
			7.6.7 Step 6: Regression Metrics
			7.6.8 Step 7: Other Regression Ensemble Models
			7.6.9 Step 8: Model R2
			7.6.10 Step 9: Explainability of the Model
		References
	Chapter 8: Composite Models: Food Security and Natural Resources
		8.1 Sugarcane Cultivation
		8.2 Energy Production
		8.3 Australia
			8.3.1 Sugarcane and Australian Economy
			8.3.2 World Sugar Prices
			8.3.3 Sugar Demand
			8.3.4 Climate Change
			8.3.5 Precipitation and Rainfall
			8.3.6 Maximum Temperature
				8.3.6.1 NOAA: Advanced Very-High-Resolution Radiometer (AVHRR) and Climate Change Parameters
			8.3.7 Vegetation Index
				8.3.7.1 Province Averaged VH
				8.3.7.2 No Noise (Smoothed) Normalized Difference Vegetation Index (SMN)
		8.4 Machine Learning Model: Australia Sugarcane Production Model
			8.4.1 Data Sources
			8.4.2 Step 1: Import Libraries
			8.4.3 Step 2: Load Australia Sugarcane Production: FAO Dataset 1991-2020
			8.4.4 Step 3: Prepare DataFrame with Weather Data (Average Temperature, Min/Max Temperatures, and Precipitation)
			8.4.5 Step 4: Concat Weather Data to Prepare the Data for Modeling
			8.4.6 Step 5: Merge Production and Weather DataFrames to Prepare the Data for Modeling
			8.4.7 Step 6: Merge Data: The Final DataFrame
			8.4.8 Step 7: Correlation Values
			8.4.9 Step 8: Correlation Matrix
			8.4.10 Step 9: Prepare Final Sugarcane DataFrame with Weather Details for Regression
			8.4.11 Step 10: Regression Metrics
			8.4.12 Step 11: Other Regressors (Ensemble)
			8.4.13 Step 12: Model Explainability
		8.5 Machine Learning Model: Queensland, Australia, Sugar Production with Climate Projections (CMPI6) and Shared Socioeconomic ...
			8.5.1 CMIP6 Projections Data and SSP Scenarios
			8.5.2 Climate Model: Temperature Change (RCP 6.0): 2006-2100
				8.5.2.1 SSP 5-8.5
				8.5.2.2 SSP 1_1.9
				8.5.2.3 SSP 3_7.0
		8.6 Machine Learning Model: Queensland, Australia, Sugar Production with the Normalized Difference Vegetation Index (NDVI) Mod...
			8.6.1 NDVI Anomaly
			8.6.2 Step 1: Load Mozambique Vegetation Health Index (VHI) DEKAD Data
			8.6.3 Step 2: Prepare Production DataFrame
			8.6.4 Step 3: Sugarcane and MODIS VHI
			8.6.5 Step 4: Merged DataFrame Correlation Matrix
			8.6.6 Step 5: Prepare Data for Regression Model
			8.6.7 Step 6: Regression Metrics
			8.6.8 Step 7: Other Regression Ensemble Model
			8.6.9 Step 8: Model R2
			8.6.10 Step 9: Explainability of the Model
		8.7 Conclusion: Sugarcane Yield and Climate Change Impact
			8.7.1 Yield Estimation and Prediction
		References
	Chapter 9: Linkage Models: Economic Key Drivers and Agricultural Production
		9.1 Linkage Models
		9.2 Farm Drivers (Exogenous)
			9.2.1 Composite Index: IGC´s Grains and Oilseeds (GOI) Price Index and GOFI
			9.2.2 Grain Transportation Cost Indicators
			9.2.3 Government Interventions and Credit to Agriculture
				9.2.3.1 The United States of America
				9.2.3.2 China
				9.2.3.3 India
				9.2.3.4 France
				9.2.3.5 History of Credit to Agriculture
					2010
					2000
					1991
			9.2.4 Trade (Import/Export) Restrictions
			9.2.5 Location and Weather
			9.2.6 Vegetation Index: Drought and Other Extreme
		9.3 Real Economic Activity
			9.3.1 Food Price Index
			9.3.2 The Dairy Price Index
				9.3.2.1 World Sugar Prices: High Food Price Time Series
			9.3.3 The USD Exchange Rate
			9.3.4 Population Growth
			9.3.5 Inflation, Consumer Prices (Annual %) (CPI)
			9.3.6 Ratio of Export to Import Agricultural Products (EXIM)
			9.3.7 Agriculture, Forestry, and Fishing, Value Added (% of GDP)
			9.3.8 Total Reserves (Includes Gold, Current US$)
		9.4 Farm Inputs
			9.4.1 Employment in Agriculture (% of Total Employment) (Modeled ILO Estimate)
			9.4.2 Rural Population (% of Total Population)
		9.5 Food Security Indicators
			9.5.1 Shock Indicators: Per Capita Food Production Variability
			9.5.2 Domestic Food Price Volatility
		9.6 Machine Learning Model: Australia Macroeconomic Drivers and Sugarcane Production Predictive Model
			9.6.1 Data Sources
			9.6.2 Sugarcane Production Model
				9.6.2.1 Employment in Agriculture (% of Total Employment) (Modeled ILO Estimate)
				9.6.2.2 Agriculture, Forestry, and Fishing, Value Added (Constant 2015 US$)
				9.6.2.3 Fertilizer Consumption (Kilograms Per Hectare of Arable Land)
				9.6.2.4 GDP Per Capita Growth (Annual %): NY.GDP.PCAP.KD.ZG
				9.6.2.5 Rural Population (% of Total Population)
				9.6.2.6 Exchange Rate (USD/AUD)
				9.6.2.7 Vegetation Index
					Weather Impact on GDP: Direct Effect
					Weather Impact on GDP: Indirect Effect
			9.6.3 Model Development
		9.7 Machine Learning Model: Myanmar Rice Production Macroeconomic Linkage Predictive Model
			9.7.1 Data Sources
			9.7.2 Model Development
		9.8 Linkage Models and Data Flow
			9.8.1 Brazil, 1984, Sugar Price Spike: Increase in Oil Prices Has Reduced Sugar in the World Market
			9.8.2 2008 Trade Restrictions by Major Rice Suppliers
			9.8.3 Weather: Food Security Data Flow
			9.8.4 El Niño Triggers Macroeconomic Waves (Albeit, Not for Faint Surfers)
		9.9 Modeling Exogenous Varialbes and Linkage Models
			9.9.1 Drought in Brazil
		References
	Chapter 10: Heuristics and Agricultural Production Modeling
		10.1 Linkage and Projection Models
			10.1.1 Available Harvest Land for Cultivation
			10.1.2 CMIP6: SSP8.5 Harvest Land
			10.1.3 CMIP6: SSP3-7.0 Harvest Land
			10.1.4 CMIP6: SSP2-4.5 Harvest Land
			10.1.5 CMIP6: SSP1-2.6 and SSP1-1.9 Harvest Land
			10.1.6 World Population, Total
			10.1.7 More Food Required in the Future: Changing Demographics
		10.2 Thailand Nakhon Sawan Rice Production LP Linkage Model (SSP5_8.5: 2080-2099)
			10.2.1 Step 1: Define Thailand Nakhon Sawan linear programming (LP) variables
			10.2.2 Step 2: Define Variable Constraints
			10.2.3 Step 3: Define Model Constraints
			10.2.4 Step 4: Define Objective Function and Solve It
			10.2.5 Step 5: Output the Model
			10.2.6 Step 6: Redefine Objective Function (Seven Percentage Reduction in Harvested Land and Production)
			10.2.7 Step 7: Define Objective Function and Solve It (with Ten Percentage Drop in Rice Production)
			10.2.8 Step 8: Define Objective Function and Solve It (with Fifteen Percentage Drop in Rice Production)
		10.3 Thailand Nakhon Sawan Rice Production LP Linkage Model (SSP3_7.0: 2080-2099)
			10.3.1 Step 1: Define Thailand Nakhon Sawan Linear Programming (LP) Variables
			10.3.2 Step 2: Define Variable Constraints
			10.3.3 Step 3: Define Model Constraints
			10.3.4 Step 4: Define Objective Function and Solve It
			10.3.5 Step 5: Output the Model
		References
Part IV: Conclusion
	Chapter 11: Future
Appendix 1: Food Security and Nutrition
	The 17 Sustainable Development Goals (SDGs)
	Food Security Monitoring System (FSMS)
	NHANES 2019-2020 Questionnaire Instruments: Food Security
	Macroeconomic Signals That Could Help Predict Economic Cycles
	List of World Development Indicators (WDI)
		Wheat Data Disappearance and End Stocks
	Food Aids
		PL-480 or Food for Peace
	The United Nations: 17 Sustainable Development Goals (SDGs)
	The Statistical Distributions of Commodity Prices in Both Real and Nominal Terms
	Poverty Thresholds for 2019 by Size of Family and Number of Related Children Under 18 Years
Appendix 2: Agriculture
	Agricultural Data Surveys
		USDA NASS
		USDA: The Foreign Agricultural Service (FAS) Reports and Databases
			World Agricultural Production
			World Markets and Trade
			Global Agricultural Information Network (GAIN)
			Production, Supply, and Distribution (PS&D) Online
			USDA and NASA Global Agricultural Monitoring (GLAM)
			Global Agricultural and Disaster Assessment System (GADAS)
			Export Sales Reporting
			Global Agricultural Trade System (GATS)
	Data Sources
		UN DATA MARTS
		Food and Agriculture Organization of the United Nations
	Conversion Factors
	National Dairy Development Board (NDDB), India
	Worldwide: Artificial Intelligence (AI) Readiness
Appendix 3: Data World
	Global Historical Climatology Network Monthly (GHCNm)
		Datasets for Brazil
	Labor Force Statistics from the Current Population Survey
		IMF Country Index Weights
		IMF Data Bulk Download
		World Bank Data
		The World Bank Development Indicators
		Food and Agriculture Organization of the United Nations
Appendix 4: US Data
	USDA Datasets
		DATA.GOV
		Wheat Data Disappearance and End Stocks
	Dollars/Bushel: Dollars/Ton Converter
	NOAA: Storm Events Database
		Storm Events Database
		Storm Data Event Table
Appendix 5: Economic Frameworks and Macroeconomics
	Macroeconomic Signals That Could Help Predict Economic Cycles
	The US Recessions
	Labor Force Statistics from the Current Population Survey
	United Nations Statistics Department (UNSD)
		The OCED: Main Economic Indicators (MEI)
	Reserve Bank of India: Handbook of Statistics on Indian Economy
	Department of Commerce
		Bureau of Economic Analysis: The United States Department of Commerce
	United Nations Data Sources
		UN DATA MARTS
		IHS Global Economy Data
		US Commodities Futures Data
		IMF Data Access to Macroeconomic and Financial Data
		IMF Country Index Weights
		IMF Data Bulk Download
		World Bank Data
		The World Bank Development Indicators
		Food and Agriculture Organization of the United Nations
		Wheat Data Disappearance and End Stocks
	Poverty Thresholds for 2019 by Size of Family and Number of Related Children Under 18 Years
	Rice Production Manual