Optimal control analysis of COVID-19 epidemic with fear effects using artificial neural networks
Year: 2026
Type: article
Abstract: Mathematical modeling is effective in understanding the transmission dynamics of epidemics and designing control strategies. However, the behavioral responses of individuals during outbreaks are often overlooked. Epidemics spread mechanisms are influenced by fear, anxiety, and behavioral changes. Motivated by these challenges, a novel epidemic model incorporating fear dynamics is proposed. A theoretical analysis is conducted to analyze the model including basic reproduction number and stability analysis. Sensitivity analysis is conducted to identify the most influential epidemiological parameters affecting disease transmission. Furthermore, optimal control theory is employed to determine cost-effective intervention strategies that minimize the number of infected individuals while reducing implementation costs. The model simulation was carried through computing tool MATLAB. In addition, an Artificial Neural Network (ANN) framework is employed using the PyTorch deep learning library for computational approximation and forecasting. The coefficient of determination $$R^2$$ was found close to unity for all compartments with low values of other error metrics. This ensures the high predictive capability and approximation accuracy of the neural network framework. The simulation results reveal that combined vaccination, prevention, and quarantine strategies substantially reduce disease transmission. Sensitivity analysis reveals that transmission parameters positively influence disease spread, whereas vaccination, quarantine, and recovery rates contribute significantly to disease mitigation. The integration of the proposed hybrid mathematical–computational framework provides an efficient and reliable tool for epidemic prediction and public health decision-making in the control of infectious diseases.
Source: Discover Public Health
Authors: Shakeel Ahmed, Iqra Nazir, Amna Areej
Institution: University of Sialkot
Language: English
Subfield: Applied Mathematics
Field: Mathematics
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7225999013
