Aiming at the characteristics and existing problems of dynamic service resource scheduling in cloud manufacturing (CMfg) environment, this paper studies the scheduling method of CMfg dynamic service resources. Firstly, the problem of optimal scheduling of dynamic service resources is studied. The mechanism of CMfg scheduling is summarised. The operation mechanism and scheduling system of CMfg scheduling are described. Secondly, from the perspective of resource allocation, the CMfg scheduling problem is assumed. The optimal scheduling model of dynamic service resources in CMfg environment is established with the goal of time, cost, quality and capability. Then, the ant optimisation algorithm (AO) is improved, and some functions in the genetic algorithm (GA) are used to optimise the objective function. A genetic-ant optimisation fusion algorithm(GA-AO) is proposed to solve the model. Finally, taking the production of a car component as an example, the algorithm is applied as an example, and compared with the general GA and AO, the model and algorithm proposed in this paper are proved to be more feasible and effective.