2023, Vol. 8, Special Issue 4

This research paper presents a comprehensive investigation into the pre-harvest forecasting of Ragi (Hill Millet) yield within the Dang district, employing a novel approach that integrates weather parameters and biometrical characters. The study acknowledges the significance of accurate yield prediction in modern agricultural practices, aiding farmers, policymakers, and the agricultural industry in optimizing resource allocation and management. The research objectives encompass the formulation of a pre harvest model that synergistically incorporates weather data and biometrical traits to proactively estimate Ragi yield. Through the application of multiple regression model, Discriminant function analysis and Composite forecast, correlations between weather parameters, biometrical characters, and yield are discerned and harnessed to construct the pre harvest model. The study revealed that discriminant function gave better result in the 42nd and 43rd SMW with minimum forecast error % even though low R² values as compared MLR models. The combined forecast obtained by forecast values based on biometrical character and weather data found good by using Model 2_42 and Model 2_43 (discriminant function).