Project Title: 201

Crop Yield Prediction and Pest Outbreak Detection Using

Satellite Imagery and IoT Sensors

Global Problem Statement

Develop an AI-driven system that analyzes satellite imagery and IoT sensor data to predict crop yields
and detect early signs of pest outbreaks. The system should help farmers make informed decisions on
irrigation, fertilization, and pest control.

Domain
Agriculture

Pre-requirements

Familiarity with remote sensing, GIS, and agricultural practices

• Experience with image processing and machine learning models • Proficiency in Python,
TensorFlow/PyTorch, and data integration techniques • Knowledge of agronomic data and pest
management

Day 1: Data Integration and Preprocessing (ETL)

Problem Statement
Acquire, clean, and integrate datasets related to ‘Crop Yield Prediction and Pest Outbreak Detection
Using Satellite Imagery and IoT Sensors’ into a single, analysis-ready master dataset using Power Query
in Excel following a star schema design.

Dataset
• Master_Agriculture_Data.xlsx (adapted for Crop Yield Prediction and Pest Outbreak Detection Using

Satellite Imagery and IoT Sensors) Expected Deliverables

1. Technical report detailing the methodology used.

2. Cleaned and processed dataset files.

3. Visualization charts or interactive dashboard pages.

 4. Code scripts (Power Query M-code/Python/SQL/Power BI) used for analysis.

Rubric
Criteria Points

Accuracy of implementation 6

Depth of analysis and insights 4

Quality of visualizations 3

Documentation and code clarity 2

Day 2: Descriptive Statistics & Exploratory Data Analysis (EDA)

Problem Statement
Perform a comprehensive exploratory analysis on the ‘Crop Yield Prediction and Pest Outbreak
Detection Using Satellite Imagery and IoT Sensors’ dataset to uncover patterns, identify outliers, and
understand distributions.

Dataset
• Master_Agriculture_Data.xlsx (adapted for Crop Yield Prediction and Pest Outbreak Detection Using

Satellite Imagery and IoT Sensors) Expected Deliverables

1. Technical report detailing the methodology used.

2. Cleaned and processed dataset files.

3. Visualization charts or interactive dashboard pages.

4. Code scripts (Power Query M-code/Python/SQL/Power BI) used for analysis.

Criteria Points

Accuracy of implementation 4

Depth of analysis and insights 3

Quality of visualizations 2

Documentation and code clarity 1

Rubric

Day 3: Customer Segmentation using RFM Analysis

Problem Statement
Segment the entities (customers/users/assets) in the ‘Crop Yield Prediction and Pest Outbreak Detection
Using Satellite Imagery and IoT Sensors’ project using RFM or similar behavioral analysis to identify key
groups.

Dataset
• Master_Agriculture_Data.xlsx (adapted for Crop Yield Prediction and Pest Outbreak Detection Using
Satellite Imagery and IoT Sensors)

Expected Deliverables

1. Technical report detailing the methodology used.

2. Cleaned and processed dataset files.

3. Visualization charts or interactive dashboard pages.

4. Code scripts (Power Query M-code/Python/SQL/Power BI) used for analysis.

Rubric

Criteria Points

Accuracy of implementation 4

Depth of analysis and insights 3

Quality of visualizations 2

Documentation and code clarity 1

Day 4: Product Performance & Sales Analysis

Problem Statement
Analyze performance metrics and key drivers for the ‘Crop Yield Prediction and Pest Outbreak
Detection Using Satellite Imagery and IoT Sensors’ project. Build a performance analysis and visualize
portfolio health.

Dataset
• Master_Agriculture_Data.xlsx (adapted for Crop Yield Prediction and Pest Outbreak Detection Using
Satellite Imagery and IoT Sensors)

Expected Deliverables

1. Technical report detailing the methodology used.

2. Cleaned and processed dataset files.

3. Visualization charts or interactive dashboard pages.

4. Code scripts (Power Query M-code/Python/SQL/Power BI) used for analysis.

Rubric

Criteria Points

Accuracy of implementation 6

Depth of analysis and insights 4

Quality of visualizations 3

Documentation and code clarity 2

Day 5: Fraud Detection – Anomaly Detection & Rule-Based Flagging

Problem Statement
Identify suspicious activities or anomalies within the ‘Crop Yield Prediction and Pest Outbreak
Detection Using Satellite Imagery and IoT Sensors’ project using rule-based flagging and statistical
detection.

Dataset
• Master_Agriculture_Data.xlsx (adapted for Crop Yield Prediction and Pest Outbreak Detection Using
Satellite Imagery and IoT Sensors)

Expected Deliverables

1. Technical report detailing the methodology used.

2. Cleaned and processed dataset files.

3. Visualization charts or interactive dashboard pages.

4. Code scripts (Power Query M-code/Python/SQL/Power BI) used for analysis.

Rubric

Criteria Points

Accuracy of implementation 4

Depth of analysis and insights 3

Quality of visualizations 2

Documentation and code clarity 1

Day 6: Key KPI Monitoring Dashboard

Problem Statement
Create a ‘Command Center’ dashboard to monitor critical KPIs for ‘Crop Yield Prediction and Pest
Outbreak Detection Using Satellite Imagery and IoT Sensors’ across operational and strategic
dimensions.

Dataset
• Master_Agriculture_Data.xlsx (adapted for Crop Yield Prediction and Pest Outbreak Detection Using
Satellite Imagery and IoT Sensors)

Expected Deliverables

1. Technical report detailing the methodology used.

2. Cleaned and processed dataset files.

3. Visualization charts or interactive dashboard pages.

4. Code scripts (Power Query M-code/Python/SQL/Power BI) used for analysis.

Rubric

Criteria Points

Accuracy of implementation 4

Depth of analysis and insights 3

Quality of visualizations 2

Documentation and code clarity 1

Day 7: Interactive Dashboard Development (Power BI / Tableau)

Problem Statement
Combine all previous analyses for ‘Crop Yield Prediction and Pest Outbreak Detection Using Satellite
Imagery and IoT Sensors’ into a single, cohesive, interactive dashboard that tells a complete story.

Dataset
• Master_Agriculture_Data.xlsx (adapted for Crop Yield Prediction and Pest Outbreak Detection Using
Satellite Imagery and IoT Sensors)

Expected Deliverables

1. Technical report detailing the methodology used.

2. Cleaned and processed dataset files.

3. Visualization charts or interactive dashboard pages.

4. Code scripts (Power Query M-code/Python/SQL/Power BI) used for analysis.

Rubric

Criteria Points

Accuracy of implementation 4

Depth of analysis and insights 3

Quality of visualizations 2

Documentation and code clarity 1

Day 8: Sales Forecasting using Time Series Analysis

Problem Statement
Use historical data from ‘Crop Yield Prediction and Pest Outbreak Detection Using Satellite Imagery and
IoT Sensors’ to forecast future trends or values for the next 90 days with confidence intervals.

Dataset
• Master_Agriculture_Data.xlsx (adapted for Crop Yield Prediction and Pest Outbreak Detection Using
Satellite Imagery and IoT Sensors)

Expected Deliverables

1. Technical report detailing the methodology used.

2. Cleaned and processed dataset files.

3. Visualization charts or interactive dashboard pages.

4. Code scripts (Power Query M-code/Python/SQL/Power BI) used for analysis.

Rubric

Criteria Points

Accuracy of implementation 4

Depth of analysis and insights 3

Quality of visualizations 2

Documentation and code clarity 1

Day 9: Advanced Analytics – Cohort & Retention Analysis

Problem Statement
Analyze retention and behavior over time using cohort analysis for the ‘Crop Yield Prediction and Pest
Outbreak Detection Using Satellite Imagery and IoT Sensors’ project to identify longterm trends.

Dataset
• Master_Agriculture_Data.xlsx (adapted for Crop Yield Prediction and Pest Outbreak Detection Using
Satellite Imagery and IoT Sensors)

Expected Deliverables

1. Technical report detailing the methodology used.

2. Cleaned and processed dataset files.

3. Visualization charts or interactive dashboard pages.

4. Code scripts (Power Query M-code/Python/SQL/Power BI) used for analysis.

 Rubric

Criteria Points

Accuracy of implementation 4

Depth of analysis and insights 3

Quality of visualizations 2

Documentation and code clarity 1

Day-wise Mark Distribution

Day Theme Marks

1 Data Integration and Preprocessing (ETL) 15

2 Descriptive Statistics & Exploratory Data Analysis (EDA) 10

3 Customer Segmentation using RFM Analysis 10

4 Product Performance & Sales Analysis 15

5 Fraud Detection – Anomaly Detection & Rule-Based Flagging 10

6 Key KPI Monitoring Dashboard 10

7 Interactive Dashboard Development (Power BI / Tableau) 10

8 Sales Forecasting using Time Series Analysis 10

9 Advanced Analytics – Cohort & Retention Analysis 10

Total 100