The cloud manufacturing system is a customer-oriented paradigm that benefits from centralized management of all available resources. This paper focuses on the integration of sub-task scheduling and logistics (ISSL) in the cloud manufacturing system with two main contributions: 1) using a combined transportation system which provides the advantage of transporting more than one sub-task by a vehicle at the same time, and 2) tasks can have different structure types including sequential or parallel. To get the model closer to reality, two factors are considered: 1) different task arrival times; and 2) The setup time/cost. The proposed model aims to optimize task completion time, cost, and average quality service concurrently. To solve the proposed model, GAMS software is utilized for small/medium-sized samples while a genetic algorithm is developed for larger-size samples. Three comparative studies are conducted; the findings show that employing combined logistics significantly impacts the cost imposed on cloud systems while the real task arrival time to the cloud platform and setup time/cost have a notable effect on the task completion time and cost, respectively. Eventually, a sensitivity analysis is undertaken to gain insight into the impact of execution time, service cost, and user preferences on the final solution.