An Improved Agent-Based Model Using Discrete Event Simulation for Nonpharmaceutical Interventions

Abstract

The traditional agent-based model requires high computing power of the central processing unit. Thus, an improved agent-based model combined with the discrete event simulation method is proposed. The result of the equation-based Susceptible-Exposed-Infective-Asymptomatic-Recovered (SEIAR) model with the same parameter combination, which has been demonstrated to be effective, is used to verify the validity of this improved agent-based model. Additionally, an analysis based on simulation results of the Contact Tracing Measure (CTM), Location-Based Checking-Testing Measure (LCTM), Lockdown Measure (LM), Mobile Cabin Isolation and Hospital Measure (MCHM) is presented. The simulation results show that implementing long-term lockdown measures has the best effect on epidemic control. Moreover, according to the simulation results, we inferred that using only nonpharmaceutical epidemic prevention measures may result in a second outbreak of COVID-19 owing to the risk of asymptomatic transmission.

Keywords

• Agent-based model
• asymptomatic
• COVID-19
• discrete event simulation