Dynamic analysis of COVID-19 model

In this thesis, the spread dynamics of the Covid-19 pandemic were investigated using mathematical modeling methods. The study first addressed the SIR and its derivatives, which are classical deterministic models, and conducted analyses of the fundamental reproduction number (R_0 ), equilibrium points, and stability of these models. Subsequently, mathematical models developed for Covid-19 were analyzed, investigating the positivity, boundedness, and invariant regions of the solution. The memory effect and long-term dependency structure of the epidemic were investigated using fractional-order differential equations. Fractional models based on Caputo, Conformal, and Grünwald-Letnikov derivatives were solved using analytical and numerical methods. In this context, the Homotopy Analysis Method (HAM), q-HATM, Variational Iteration Method (VIM), and Non-Standard Finite Difference (NSFD) schemes were applied, and the resulting solutions were comparatively evaluated. Additionally, the uncertain spread of Covid-19 was modeled using stochastic differential equations, and numerical solutions were obtained using the Euler–Maruyama, Milstein, and Runge–Kutta methods. Furthermore, time-delay differential equations were used to incorporate effects such as immunity and incubation period into the model. The results demonstrate that the spread of Covid-19 can be more realistically explained not only with deterministic models but also with fractional-order and stochastic models. In this respect, the thesis contributes to both the epidemiological modeling literature and the applications of fractional analysis and numerical methods.