LAUSR

To improve the accuracy of service modelling and optimal selection in cloud manufacturing (CMfg), a multi-level modelling methodology is proposed to describe manufacturing services. In this methodology, manufacturing services are divided into three levels: resource, functional and process services. Based on time, cost and reputation analysis of these three service levels, the corresponding objective functions and services composition constraints are established. Considering intelligent optimal selection, a niching behaviour-based gravitational search algorithm (NGSA) is designed to address manufacturing service composition and optimal selection (MSCOS) problems. In NGSA, the niche crowding factor and fitness sharing technology are introduced to the standard gravitational search algorithm (GSA) to improve its convergence speed and accuracy. The results of a simulation experiment demonstrate that the proposed algorithm can find better solutions in less time than previous algorithms, such as the adaptive genetic algorithm (AGA) and the modified particle swarm optimization (MPSO) algorithm.