Parallel Machine With Different Address Scheduling Problem Under Cloud Manufacturing

This paper considers parallel machine with different address scheduling problems under cloud manufacturing. The rapid development and wide application of science and technology, the Internet of things, cloud computing and big data, have a huge impact on the production and operations management of modern-day manufacturing industry. It is convenient for remote enterprises or customers to trade the right of use of manufacturing resources in real time via the Internet. This has given birth to cloud manufacturing. There may be great differences of position of manufacturing resources through the cloud manufacturing platform. So the selection and the address of machine have significant effect on delivery in scheduling.In the paper, we firstly consider a special case, where there is a distribution center between manufacturers and customers under cloud manufacturing. Our goal is to find a schedule that minimizes the service span among all feasible schedules, so as to improve customer satisfaction. We give an approximation algorithm, named MB algorithm and show our algorithm has worst-case performance ratio at most 2. At last, a large number of random computational experiments are performed to test the average-case performance, and our results indicate that the performance of the algorithms is quite satisfactory.Then, we consider parallel machine with different address scheduling problems under cloud manufacturing, where there is no distribution center, and delivery times are dependent of the machines. In this problem, we have to consider delivery times. Our goal is also to find a schedule that minimizes the servicespan, which is the maximum service time among all the jobs. We give two deterministic algorithms, SDT algorithm and MSDT algorithm; construct a meta-heuristic based on simulated annealing. The performances of the heuristics are evaluated empirically by running them through large sets of random data. Our results show that MSDT algorithm outperforms the SDT algorithm by a wide margin, especially when the delivery times are large. The simulated annealing algorithm outperforms both deterministic algorithms, and it can effectively solve the problem in a reasonable amount time.