A Study on Parameter Sensitivity Analysis of the Virus Spread Optimization

Abstract

The virus spread optimization (VSO) is a radically new metaheuristic optimization algorithm to mimic the viral behavior and spread of viruses for continuous optimization. Due to the multiple search strategies design, the VSO achieves an excellent performance on a series of well-known benchmark functions in terms of the solution quality, convergence rate and stability. Yet the number of control parameters involved in the VSO algorithm is relatively larger than those of other popular metaheuristics such as genetic algorithm (GA) and particle swarm optimization (PSO). Besides, there is rarely any study on the possible impact of such parameters on the performance of the VSO as based on the default parameter settings when compared to those of other metaheuristics. In this work, the parameter sensitivity of the VSO is carefully examined by performing a suite of experiments. More importantly, the rules of thumb for the parameter tuning of the VSO is also considered. Essentially, this work reveals the impact of the parameters contributing to the success of the VSO to boost the research of this promising optimization algorithm.