Prediction of an epidemic spread based on the adaptive genetic algorithm

Abstract

In recent years, coronavirus disease 2019 (COVID-19) has plagued the world, causing huge losses to the lives and property of people worldwide. How to simulate the spread of an epidemic with a reasonable mathematical model and then use it to analyze and to predict its development trend has attracted the attention of scholars from different fields. Based on the susceptible–infected–recovered (SIR) propagation model, this work proposes the susceptible–exposed–infected–recovered–dead (SEIRD) model by introducing a specific medium having many changes such as the self-healing rate, lethality rate, and re-positive rate, considering the possibility of virus propagation through objects. Finally, this work simulates and analyzes the propagation process of nodes in different states within this model, and compares the model prediction results optimized by the adaptive genetic algorithm with the real data. The experimental results show that the susceptible–exposed–infected–recovered–dead model can effectively reflect the real epidemic spreading process and provide theoretical support for the relevant prevention and control departments.