Research on module partition of complex products considering influence of change propagation

Abstract

In the context of digitalization, the rapidly changing market and ever-improving technological environment are driving the propagation of design changes, which have become a major source of risk for the successful development of complex products (CPs). To quickly respond to changing demands and improve the efficiency of CPs development, this study aimed to reveal and express the intrinsic mechanism of change propagation and develop a product module partition method that considers the influence of change propagation. First, a multilayer weighted network of CPs was constructed based on the Function–Behavior–Structure (FBS) model. Subsequently, interval-valued intuitionistic fuzzy sets (IVIFS) were introduced to calculate the FBS correlation strength, and a dynamic weighting method was proposed based on IVIFS accuracy and piecewise functions. The Louvain algorithm was proposed to solve community detection problems in directed weighted networks. Finally, the viability and effectiveness of the proposed method were verified using a centrifugal unit as an example and comparing with the modified GN algorithm.