This paper aims to minimize makespan for the hybrid flowshop scheduling problem. We present a novel Fruit Fly Optimization (FFO), called multi-swarm FFO, by introducing a multiple-swarm strategy and a competition-and-updating mechanism to the basic FFO. The parameters and operators for the presented MMFO algorithm are calibrated by means of a design of experiments approach. The numerical comparisons show that MFFO performs much better than several well-known metaheuristics in the literature for the considered HFS problem.