Cloud manufacturing (CMfg) is an emerged manufacturing paradigm oriented towards services via a network. In which, distributed manufacturing capabilities and resources have been collaborating together through CMfg platform to fulfill the required tasks in a perfect way. In this research, firstly, an architecture of the CMfg environment was proposed for machining complex parts and assembly lines. Then, a genetic algorithm was developed to accommodate the complexities involved in CMfg problems, represented in its larger size, diversity of tasks and dynamism. In addition, two strategies have been proposed for integration with the genetic algorithm as a means of improving its efficiency. The experimental results indicate that merging these strategies into the genetic algorithm is effective in scheduling multiple tasks on a large number of resources in terms of improving multiple objectives such as makespan and the cost of transportation for the entire manufacturing life cycle.