Today there is a line of research-oriented to the design of algorithms inspired by nature. Many of these algorithms work in continuous spaces. On the other hand, there is a great amount of combinatorial optimization problems (COP) which have application in the industry. The adaptation of these continuous algorithms to resolve COP is of great interest in the area of computer science. In this article we apply the percentile concept to perform the binary adaptation of the Sine-Cosine algorithm. To evaluate the results of this adaptation we will use the set covering problem (SCP). The experiments are designed with the objective of demonstrating the usefulness of the percentile concept in binarization. In addition, we verify the effectiveness of our algorithm through reference instances. The results indicate that the binary Percentile Sine-Cosine Optimization Algorithm (BPSCOA) obtains adequate results when evaluated with a combinatorial problem such as the SCP.