A multi-population genetic algorithm is presented for Job Shop scheduling problem targeting to minimize the makespan.The coding scheme of chromosomes is defined based on the process constraints

genetic operators are defined according to job operations

so all chromosomes are feasible.By the competition and sharing mechanism among populations

the computing resource is utilized more efficiently

and prematurity is overcome effectively.These improve the convergence of the multi-population genetic algorithm.Experiments on Job Shop Benchmarks show that this algorithm has good performance.