Leveraging predictive models for accurate crop yield forecasting-smart harvest approach
5 / 0
Authors
-
Gunjan Chauhan
Student, Associate Professor AMITY, Amity University, Noida Uttar Pradesh
-
Sanjeev Tomar
Student, Associate Professor AMITY, Amity University, Noida Uttar Pradesh
-
Anshika Panwar
Student BVCE Delhi, Amity University, Noida Uttar Pradesh
-
Anushree Misra
ASV USA, Amity University, Noida Uttar Pradesh
-
Ranjit Kumar Paul
Senior Scientist IASRI, Amity University, Noida Uttar Pradesh
-
Md Yeasin
Senior Scientist IASRI, Amity University, Noida Uttar Pradesh
-
Anil Kumar
ADG Coordination, ICAR, Amity University, Noida Uttar Pradesh
-
Sanjeev Panwar
Principal Scientist ICAR, Amity University, Noida Uttar Pradesh
Keywords:
Crop yield forecasting, predictive analytics, machine learning, artificial neural networks (ANN), time series prediction, smart agricultureAbstract
This research introduces Smart Harvest, a data-driven forecasting system developed to predict the yield of key crops such as sugarcane, wheat, rice, and maize. By combining historical yield records with rainfall statistics, the model captures the intricate relationship between crop production and climatic variation. The study evaluates several predictive techniques, including ARIMA, ARIMAX, Support Vector Regression (SVR), Random Forest, and Artificial Neural Networks (ANN). Through a comparative performance analysis using metrics like RMSE, MAE, and MAPE, the results highlight the effectiveness of rainfall-integrated models in improving forecasting accuracy. The findings provide a robust foundation for enabling timely, informed agricultural planning and risk management strategies.
