Predictive modeling of pest spread in tea plants using an intelligent computational approach.

Tea is a popular beverage prepared from the leaves of the Camellia sinensis plant, embraced by a good share of the world's population. However, pests and predators threaten its production, which must be controlled without adversely affecting the natural resources. This research presents a mathematical framework for a predatory tri-trophic system that captures the growth of tea plants, pests, and predators. Bayesian Regularization Backpropagation Neural Network (BR-BNN), an intelligent computational approach, is used to derive the output solutions of the presented mathematical model. The Fourth Order Runge-Kutta Method (RKM-4) is utilized to find the target solutions, and different scenarios and parameter variations are employed to determine the model's solutions and assess the stability of BR-BNN. Furthermore, the comparison graphs of BR-BNN solutions with the target solutions also exhibit how well the BR-BNN performs with a minimal percentage of error in all the results. The findings can also help address challenges related to increasing yield, protecting the environment, and increasing the sustainability of tea production.