The Set Covering Problem (SCP) is a matrix that is composed of zeros and ones and consists in finding a subset of zeros and ones also, in order to obtain the maximum coverage of necessities with a minimal possible cost. In this world, it is possible to find many practical applications of this problem such as installation of emergency services, communications, bus stops, railways, airline crew scheduling, logical analysis of data or rolling production lines. SCP has been solved before with different nature inspired algorithms like fruit fly optimization algorithm. Therefore, as many other nature inspired metaheuristics which imitate the behavior of population of animals or insects, Artificial Fish Swarm Algorithm (AFSA) is not the exception. Although, it has been tested on knapsack problem before, the objective of this paper is to show the performance and test the binary version of AFSA applied to SCP, with its main steps in order to obtain good solutions. As AFSA imitates a behavior of a population, the main purpose of this algorithm is to make a simulation of the behavior of fish shoal inside water and it uses the population as points in space to represent the position of fish in the shoal.