The Manufacturing Cell Design Problem (MCDP) consists on the grouping of machines, from a production plant, into independent manufacturing cells. The idea is that each cell processes families of parts, thus reducing movements of the parts among cells. The main reason to perform this procedure is to reduce cost and time in the production of parts. To solve this problem, we use the Spider Monkey Optimization metaheuristic (SMO), an algorithm based on nature that has a fission-fusion social structure. This has resolved 90 instances of the problem of manufacturing cells, taken from the experiments of Boctor. The results obtained have shown that the SMO algorithm is able to successfully reach all the optimal values.