Mathematical Modeling and Control of Invasive Species

Using Logistic Regression and Optimization

Presenter(s)

Maheshi Galahitiyawe Walawwe

Abstract or Description

Mathematical Modeling and Control of Invasive Species Using Logistic Regression and Optimization

Invasive species harm the environment and economy by taking over natural habitats and competing with native plants and animals. This study explores mathematical modeling approaches for predicting and controlling invasive species spread, with a focus on Japanese honeysuckle. We begin by examining an existing model that uses state transition relations and logistic growth functions to describe population dynamics within a structured grid. The model incorporates factors such as carrying capacity, environmental variables, and spatial interactions between neighboring cells. Building upon the existing framework, we propose an improved model that refines the estimation of carrying capacity () using logistic regression. This approach integrates key ecological variables, including soil type, altitude, climate conditions, and land disturbances, allowing more precise determination of species proliferation potential. Furthermore, we introduce a negative exponential dispersal kernel to model propagule movement between adjacent cells, improving the accuracy of species spread predictions. As a part of future work, we aim to optimize species management by refining an optimization model that minimizes and controls the species spread. Our finding will contribute to the development of adaptive management plans by incorporating real-world environmental data and advanced computational techniques, ultimately improving conservation efforts and habitat restoration strategies.

Mentor

Dr. Lakmali Weerasena