Dynamic learning super network modeling of a complex product system based on multi-organization cooperation

Abstract

Super network modeling has become an effective approach for analyzing complex systems. In this study, a super network is proposed in terms of systems science. A vital aspect of the modeling is the unique dynamic mechanism of the complex system, particularly the preferential mechanism. The preferential mechanism is the driving force of system evolution. However, in current studies, preferential probability has been mostly related to node degree, and there has been little consideration of multi-attribute decision-making based on complex system characteristics (multi-level, multi-traffic, etc.). Furthermore, association analysis of driving forces and the topology structure of the complex system should be highlighted to explore the operating mechanism. In this study, we consider a complex production system (CoPS) as the research object, propose a unique learning motivation mechanism and interaction mechanism of the CoPS, develop a preferential algorithm based on the interval-valued intuitionistic uncertain linguistic (IVIUL) operator, and construct a multi-organization knowledge learning super network model. A simulation experiment was conducted to explore the effect of the preferential parameter on learning performance. The results show that the feature of the project team has an important influence on the learning improvement velocity of the super network.