Abstract: Considering the inventory cost of workpieces in process, machining time cost, direct labor cost, inventory cost of early finished products and cost increment due to late finished products, a method calculating production cost of dual-resource(machine tool plus labor) flexible job shop scheduling is proposed. A new hybrid genetic algorithm is designed by embedding simulated annealing algorithm (SA) into genetic algorithm (GA), where GA is used to search for a group of better solutions to the problem of minimizing production cost and then SA is applied to searching them for the best one. Introducing an operation-based encoding and a new decoding method, several kinds of crossover operations are used to enable the algorithm to search the optimal solution in solvable space as far as possible. To avoid missing the optimal solution during the course of evolution, the optimized solution in every generation is kept then updated uninterruptedly. An example is given to prove that the scheduling method is feasible and efficient.