Project Description

Crop yield depends heavily on the weather. This relationship is being shaped by growing empirical research to develop the consequences of climate change on the sectors. Saudi Arabia maize production was 89 thousand tonnes in 2019, up from 84 thousand tonnes the previous year. Forecasting maize yield with some lead time can help growers plan for need and often limited human resources and assistance in strategic business decisions. The primary goal of this proposal is to demonstrate the correlation among the various weather parameters closely related to maize yield and to estimate forecasts by using principal component analysis and machine learning techniques, namely single-layer neural network (NN) and genetic random forest. Basically, the machine learning technique provides improved skills in forecasting maize yields, when compared to principal component analysis (PCA). Further, the neural network offers most of the potential skills in predicting maize yield. While analysis of one growing season it has capable of forecasting crop yields with reasonable skills, more efforts are required to estimate this approach in various fields in the regions of Saudi Arabia.