Predictive modeling of pest spread in tea plants using an intelligent computational approach
Hamad Jan1, Muhammad Sulaiman1, Muhammad Fawad Khan2
1Department of Mathematics, FP & NS, AWKUM, Mardan, 23200, Pakistan.
Scientific Reports
|June 4, 2026
Summary
This study introduces a mathematical model for tea plant, pest, and predator interactions. It uses a neural network approach to find solutions, aiding sustainable tea production and environmental protection.
Area of Science:
- Ecology
- Mathematical Biology
- Agricultural Science
Background:
- Tea production faces threats from pests and predators, necessitating sustainable control methods.
- Ecological balance is crucial for maintaining natural resources in agriculture.
Purpose of the Study:
- To develop a mathematical framework for a tri-trophic system involving tea plants, pests, and predators.
- To apply computational intelligence for analyzing ecological dynamics in tea cultivation.
Main Methods:
- A mathematical model representing a predatory tri-trophic system was formulated.
- Bayesian Regularization Backpropagation Neural Network (BR-BNN) was employed to solve the model.
- The Fourth Order Runge-Kutta Method (RKM-4) was used for validation and comparison.
Main Results:
- BR-BNN accurately derived solutions for the mathematical model with minimal error.
- Parameter variations and scenario analyses confirmed the model's stability and effectiveness.
- Comparison graphs demonstrated high concordance between BR-BNN and RKM-4 solutions.
Conclusions:
- The study validates BR-BNN as a reliable computational tool for ecological modeling in agriculture.
- The findings support sustainable tea production by providing insights into pest and predator dynamics.
- This research contributes to optimizing tea yield while safeguarding environmental health.
