Development of a Machine Learning-Based Prediction Model for Sugarcane Yield: A Study with Special Reference to Meerut, Uttar Pradesh, India

Abstract

Sugarcane is a vital cash crop in India, with Uttar Pradesh being one of the leading producing states. Accurate yield prediction is essential for optimizing agricultural planning, resource allocation, and market strategies. Traditional statistical models often fail to capture the complex, non-linear interactions between climatic and soil parameters affecting yield. This study leverages machine learning (ML) techniques to develop a robust prediction model for sugarcane yield in Meerut, Uttar Pradesh, using historical climatic and soil data from 2013 to 2023. Four ML models—Multiple Linear Regression (MLR), Support Vector Regression (SVR), Random Forest Regression (RFR), and Artificial Neural Networks (ANN)—were implemented and compared based on Mean Squared Error (MSE) and R² score. The results indicate that ANN achieved the highest prediction accuracy (R² = 0.94), followed by SVR and RFR, while MLR performed comparatively lower. Sensitivity analysis revealed that solar radiation and soil moisture significantly impact yield outcomes. This study highlights the potential of AI-driven agricultural forecasting in precision farming, aiding farmers and policymakers in improving yield management. The findings emphasize the need for integrating ML models into agricultural practices for sustainable crop production and resource efficiency.