electricsheepafrica/african-agro-processing-value-add

---
language:
- en
license: apache-2.0
task_categories:
- tabular-regression
- tabular-classification
- other
pretty_name: African Agro-Processing Value Addition Dataset
tags:
- agriculture
- agro-processing
- value-addition
- africa
- agritech
- private-equity
- manufacturing
- supply-chain
- export-readiness
- value-chain
- food-processing
- emerging-markets
size_categories:
- 100K<n<1M
configs:
- config_name: baseline
data_files:
- split: train
path: african_agro_processing_baseline.csv
description: >
Baseline scenario reflecting current agro-processing conditions across 15 African countries.
Represents status quo conversion rates, value addition margins, capacity utilization (~65%),
and technology adoption levels (~45%). Calibrated against FAO and World Bank 2023 benchmarks.
- config_name: value_chain_integration
data_files:
- split: train
path: african_agro_processing_value_chain_integration.csv
description: >
Value chain integration scenario modeling improved coordination between upstream suppliers
and downstream processors. Features 12% higher conversion rates, 25% higher value addition
margins, 18% lower processing losses, and 35% higher by-product utilization. Based on
UNIDO value chain integration studies and AfDB agro-industrialization programs.
- config_name: export_oriented
data_files:
- split: train
path: african_agro_processing_export_oriented.csv
description: >
Export-oriented scenario optimized for international market compliance and competitiveness.
Features 40% higher value addition margins, 25% lower processing losses, 50% higher
by-product utilization, 85% capacity utilization, and 78% technology adoption. Aligned
with Afreximbank export financing criteria and international quality standards.
- config_name: combined
data_files:
- split: train
path: african_agro_processing_combined.csv
description: >
Combined dataset containing all three scenarios (baseline, value_chain_integration,
export_oriented) for comparative analysis and scenario modeling.
---

# African Agro-Processing Value Addition Dataset

A comprehensive dataset of agro-processing value chain metrics across 15 African countries, designed for AgriTech investors, private equity analysts, development finance institutions, and policy researchers.

## Dataset Overview

| Metric | Value |
|--------|-------|
| **Total Records** | 183,096 |
| **Countries** | 15 |
| **Product Categories** | 15 |
| **Scenarios** | 3 (baseline, value_chain_integration, export_oriented) |
| **Columns** | 59 |
| **Time Period** | 2019-2024 (monthly granularity) |
| **Firms Simulated** | ~2,500+ unique processing facilities |

## Coverage

### Countries
South Africa, Nigeria, Kenya, Egypt, Morocco, Ghana, Ethiopia, Tanzania, Uganda, Ivory Coast, Senegal, Cameroon, Zambia, Mozambique, Rwanda

### Product Categories
Coffee, Cocoa, Tea, Maize, Cassava, Rice, Sugarcane, Cotton, Palm Oil, Dairy, Meat Processing, Edible Oils, Beverages, Spices, Horticulture

## Scenarios

### Baseline
Current-state agro-processing operations reflecting existing infrastructure, technology adoption, and market conditions across Africa. Calibrated against:
- FAO Agro-Processing Industry Reports (2020-2024)
- World Bank Africa Agro-Processing Competitiveness data (2023)

### Value Chain Integration
Models the impact of improved coordination between input suppliers, processors, and distributors. Parameters based on:
- UNIDO Agro-Processing Value Chain Analysis (2022-2024)
- AfDB Agro-Industrialization in Africa studies (2023)

### Export Oriented
Optimized for international market access with higher quality standards, certification rates, and export readiness. Calibrated against:
- Afreximbank Agro-Processing Financing Reports (2023)
- International quality standards (HACCP, ISO 22000, Organic)

## Key Metrics Tracked

### Production Metrics
- Raw commodity input volumes (tons)
- Processing conversion rates
- Processed output volumes
- Processing loss rates and volumes
- By-product utilization rates and volumes

### Value Addition
- Value addition margins (multiplier)
- Quality grade distribution (premium, grade 1-3, below grade)
- Grade premium multipliers
- Value addition ratios
- Raw and processed commodity prices (USD/ton)

### Operational Efficiency
- Capacity utilization (%)
- Technology adoption level (1-5 scale)
- Labor productivity (tons/worker/month)
- Energy cost per ton (USD)
- Energy intensity (kWh/ton)

### Market & Trade
- Export readiness score (0-100)
- Cold chain compliance (%)
- Export share vs domestic share
- Transport cost per ton (USD)
- Days to nearest port

### Compliance & Certification
- Certification status (general, organic, HACCP, ISO 22000)
- Wastewater treatment rate
- Water usage (m3)

### Financial
- Raw input value (USD)
- Processed output value (USD)
- Value added (USD)
- Access to credit
- Credit amounts and interest rates
- CAPEX investment (USD)

### Country Context
- Infrastructure index
- Electricity reliability
- Logistics quality index
- Trade facilitation index
- Regulatory quality index
- Skills development index
- Processing maturity index
- Labor and energy costs

### Social
- Number of employees
- Female employment share

## Literature Calibration

All parameters are calibrated against peer-reviewed sources and institutional reports:

| Parameter | Source |
|-----------|--------|
| Conversion rates | FAO Agro-Processing Reports, UNIDO Value Chain Analysis |
| Value addition margins | AfDB Agro-Industrialization Studies, World Bank Competitiveness Data |
| Processing loss rates | FAO Post-Harvest Loss assessments, UNIDO |
| Energy intensity | World Bank Enterprise Surveys, AfDB Infrastructure Reports |
| Labor productivity | ILO Africa Employment Reports, World Bank |
| Export readiness | Afreximbank Trade Finance Reports, WTO Trade Facilitation |
| Certification rates | ITC Standards Map, UNIDO Quality Infrastructure |
| Country infrastructure | AfDB African Economic Outlook, World Bank LPI |
| Cold chain coverage | FAO Cold Chain Assessments, AfDB |
| By-product utilization | UNIDO Circular Economy in Agro-Processing |

## Usage

### Python

```python
import pandas as pd

# Load a specific scenario
baseline = pd.read_csv("african_agro_processing_baseline.csv")

# Load combined dataset
df = pd.read_csv("african_agro_processing_combined.csv")

# Filter by country and product
kenya_coffee = df[(df["country"] == "Kenya") & (df["product_category"] == "coffee")]

# Compare scenarios
for scenario in df["scenario"].unique():
subset = df[df["scenario"] == scenario]
print(f"{scenario}: Mean value addition ratio = {subset['value_addition_ratio'].mean():.3f}")
```

### HuggingFace Datasets

```python
from datasets import load_dataset

# Load specific config
ds = load_dataset("electricsheepafrica/african-agro-processing-value-add", "baseline")

# Load combined
ds = load_dataset("electricsheepafrica/african-agro-processing-value-add", "combined")
```

## File Structure

```
african-agro-processing-value-add/
├── african_agro_processing_baseline.csv (64,386 records)
├── african_agro_processing_value_chain_integration.csv (68,346 records)
├── african_agro_processing_export_oriented.csv (50,364 records)
├── african_agro_processing_combined.csv (183,096 records)
├── generate_dataset.py # Dataset generation script
├── validate_dataset.py # Validation script
├── requirements.txt # Python dependencies
└── README.md # This file
```

## Validation

Run the validation script to verify data integrity:

```bash
pip install -r requirements.txt
python validate_dataset.py
```

All 58 expected columns pass range checks for:
- Conversion rates (0.05-0.95)
- Processing loss rates (0.01-0.35)
- By-product utilization (0.0-0.95)
- Value addition margins (0.1-5.0)
- Export readiness scores (5-100)
- Cold chain compliance (0-100)
- Energy costs, labor productivity, capacity utilization
- Technology adoption levels (1-5)
- All country and product indices

## Citation

If you use this dataset in your research or analysis, please cite:

```
Electric Sheep Africa. (2024). African Agro-Processing Value Addition Dataset.
Parameters calibrated against FAO, AfDB, UNIDO, Afreximbank, and World Bank reports.
```

## License

Apache-2.0

## Disclaimer

This dataset is synthetically generated using literature-calibrated parameters for research and analytical purposes. While parameters are grounded in published institutional data, individual records do not represent specific real-world firms. Users should validate findings against primary data before making investment or policy decisions.