Random Forest–based Predictive Framework for Crop Selection and Irrigation Management

Irregular rainfall, unpredictable weather patterns, and declining soil fertility make crop selection and irrigation planning increasingly challenging for farmers. This study presents a data-driven decision-support system based on the Random Forest algorithm to recommend suitable crops and estimate weekly irrigation requirements. The model evaluates multiple field parameters, including soil moisture, temperature, humidity, rainfall, light intensity, and nitrogen–phosphorus potassium (NPK) concentrations. A curated dataset was used for training and validation, and the proposed model achieved an accuracy of 94.87%, outperforming baseline classifiers by an average margin of 8–12%. A web interface built using the Flask framework allows users to input field conditions and receive crop recommendations, confidence scores, fertilizer guidance, and irrigation estimates. The results demonstrate the effectiveness of combining machine learning with environmental sensing to support precision agriculture and promote efficient water management.