Multi -source design change propagation path optimisation based on the multi-view complex network model

Abstract

The evolution of customer requirements and key technologies makes design change unavoidable. Because of the functional and physical relations among parts and components, design change will propagate from one part to another. With the increase in product complexity, design change will always be derived from multiple sources, which is rarely studied, and monetary and time losses can be caused without effective control of multi-source changes. The ability to predict change propagation paths, and evaluate the impact early in the design process will enable engineers to make informed decisions regarding change implementation. This paper proposes a design change routing (DCR) approach based on a multi-view complex network model for the multi-source design change. First, single-view network models are built considering the functions and physical relations. A difference index between these single-view network models is defined and used in a weighted least square method to optimise the weights of each single-view network model to obtain the multi-view network model. Then, a multi-objective model is constructed to establish a DCR with multiple sources considering the cost, design change intensity, and path length. The non-dominated sorting genetic algorithm II (NSGA-II) with a parallel search strategy is developed to solve the model.