A parallel hybrid genetic algorithm for solving a kind of non-identical parallel machine scheduling problems

Aimed to a kind of non-identical parallel machine scheduling problems for minimizing the make-span, the article suggests a parallel hybrid genetic algorithm and implements it under the environment of MPI. To fulfil dynamic scheduling, the inserting of urgent task is considered by providing an array which records current machine loads. The computational results suggest that the genetic algorithm is efficient and fit for large-scale parallel machine scheduling problems and advantageous over heuristic algorithms.