The objective of this study is the resolution of a combinatorial optimization problem. This problem is the Hybrid Flow Shop (HFS) scheduling. This resolution is made by using the artificial immune algorithm sequential and parallel. We proposed an implementation of the parallel algorithm with migration. The basic idea of this method is that the initial population is divided into a number of subpopulations, multiple threads are launched and everyone is running the sequential algorithm. From time to time, individuals are exchanged between the different subpopulations (migration). An experimental study is made to show the influence of different parameters. These parameters are the choice of replacement strategy, the number of subpopulations and migration frequency. A comparison between sequential and parallel version is provided. The results of this comparison confirmed that the parallel model improves the solutions obtained in terms of quality.