A multi-objective service composition method considering the interests of tri-stakeholders in cloud manufacturing based on an enhanced jellyfish search optimizer

Abstract

Service composition (SC) plays an important role in cloud manufacturing (CMfg), and the comprehensive qualities of CMfg tasks depend on the solution quality of SC largely. The current SC models are mostly customer-centric, which ignore to a great extent the interests of CMfg service platform (CMSP) and cloud service providers (CSPs). However, all of the three roles involved are indispensable in the CMfg system, only by taking all their interests into account simultaneously can the CMfg system get stable development and extensive use. To this end, after analyzing the core interests of the tri-stakeholders, a three-objective SC (TOSC) model is proposed considering the quality of service (QoS), sustainability and cooperativity at the same time in this paper. Then, an enhanced multi-objective jellyfish search (EMOJS) is developed to solve the complex TOSC model efficiently, in which four improvement strategies are designed to enhance its search capacity, respectively are the elitist improvement by Cauchy mutation, the modification for position update, the adjustment for time control mechanism and the interspecific synergistic mechanism. Finally, 30 benchmark functions and TOSC problems with different scales are used to verify the effectiveness of the proposed model and algorithm. Experiment results suggest that the TOSC model can well balance the interests of tri-stakeholders in CMfg and the EMOJS is superior to the common-used algorithms in overall performance.