AgriVision Yield Analytics
Completed PipelineAn end-to-end agricultural intelligence and crop yield forecasting platform designed to support climate resilience, food security, and data-driven farm planning.
“Transform fragmented agronomic, environmental, and historical harvest data into reliable crop yield predictions and actionable operational insights for farmers and agricultural planners.”
Business Problem & Objectives
Transform fragmented agronomic, environmental, and historical harvest data into reliable crop yield predictions and actionable operational insights for farmers and agricultural planners.
Ingested Data Inputs
Kaggle Dataset covering crop production metrics, rainfall patterns, temperature anomalies, fertilizer usage, and regional cultivation data.
Source: Kaggle Dataset
Data Pipeline Architecture
Machine Learning & Statistical Method
Built on Databricks using PySpark and Delta Lake. Trained an MLlib Random Forest regression model to forecast expected crop yields across varying environmental and input conditions. Evaluates feature importance to identify which agronomic and climatic variables have the strongest impact on harvest outcomes, allowing stakeholders to simulate "what-if" scenarios.
Decision Engine & User Interface
Interactive Streamlit Web App: Provides an accessible front-end interface where users can input farm parameters, explore regional yield trends, and generate real-time yield forecasts without touching code. Strategic Relevance: Directly addresses food security and agricultural innovation challenges in Botswana by enabling proactive resource allocation, mitigating climate-driven harvest risks, and modernizing agricultural decision-making.
Analytical Scope: Independent prototype built on public data while preparing to take on client work.
Independent prototype built on public data while we prepare to take on client work. Not produced for or endorsed by any organisation.