ANN for prediction of Area and Production of Maize crop for Upper Brahmaputra Valley Zone of Assam

It is very important to have accurate, reliable and timely information on crop Area, crop Production and land use for making certain important decisions by the planners and policy makers for the development of agriculture. The present study is carried out to predict the crop Area and crop Production (Maize) of Upper Brahmaputra Valley Zone of Assam using Artificial Neural Networks (ANNs). Multilayer Perceptron (MLP) with single hidden layer and Radial Basis Function (RBF) network have been trained with the secondary data of the Area, Maize Production and meteorological data obtained from various sources. The appropriate model for each of the network is identified. The performance of the developed ANN models has been measured using Root Mean Squared Errors (RMSE) and Correlation Coefficients (CC). The accuracy of the developed ANN models has been compared with Multiple Linear Regression (MLR) Model. The experimental results show that MLP and RBF models outperform MLR model. Sensitivity analysis has been performed for Prediction of Maize Production and results show that temperature (maximum) is the most sensitive parameter for Maize Production followed by technology index for Upper Brahmaputra Valley Zone of Assam.