Fractional vs classical modeling of monkeypox transmission with risk perception dynamics: A comparative study

Understanding the interplay between biological memory and human behavioral responses is essential for accurately characterizing monkeypox transmission dynamics. In this study, we propose a fractional-order epidemiological model that integrates memory-dependent processes and behavioral responses to understand and predict monkeypox outbreaks. We estimate model parameters from reported monkeypox incidence data of Democratic Republic of the Congo using maximum likelihood estimation, and parameter uncertainty is quantified through bootstrap resampling with a negative binomial error structure. The fractional-order model is compared with its classical integer-order counterpart using goodness-of-fit, model selection criteria, and predictive performance on training and testing datasets. Results show that the fractional-order model provides a more accurate and robust description of the observed epidemic dynamics and consistently achieves lower prediction errors. Numerical simulations indicate that increased risk perception can effectively suppress disease transmission. Sobol global sensitivity analysis further identifies the parameters that most strongly influence disease dynamics. These results demonstrate that incorporating memory effects and behavioral responses improves the predictive capability of mathematical models for monkeypox transmission and provides insights for designing effective control strategies.