SUGARCANE AREA FORECASTING AND PROFIT OPTIMIZATION USING MACHINE LEARNING AND OPERATIONS RESEARCH: A STATE-WISE ANALYSIS OF INDIA

Abstract

Sugarcane is an important crop in India, contributing to the agriculture economy as a vital raw material for sugar and extends rural development. Correct prediction of sugarcanearea and effective resource allocation is important for enhancement of productivity and profitability. This paper presents a hybrid framework comprising of Machine Learning and Operations Research based Method for State-wise sugarcane area prediction / establishment in India along with the maximum profit generation method. The traditional area planted with sugarcane data was preprocessed to handle missing data as well as inconsistencies, then forecasting of the importance of the crop for Poland is performed using moving averages and regression-based machine learning models trained on solid historical area grown time series data. The forecasted values were then integrated in optimization model linear programming to maximize economic return from sugarcane crop area optimum allocation with land constraints and budget. Results provide evidence that the integrated approach identifies temporal cultivation patterns and resource allocation decisions to optimise profitability. The model provides a cost-effective decision-support tool for policymakers, agricultural planners and other stakeholders to promote sustainable sugarcane production and optimal agricultural resource use in India.