Product Change Propagation with Complex Network for Multiple Change Sources

In the production process of products, there often are new requirements for technology, performance and others. Meanwhile, accompanied the constraints of spare parts supply. As a result, products often need to be redesigned. There have been a number of studies on product change propagation. However, there has been a lack of estimation of redesigned product parameters. Especially in the case of multiple change sources in the product, which is more difficult to estimate the product redesign parameters. In this paper, based on qualitative dependence analysis and quantitative dependence analysis, the dependence of parts and components of the product can be identified. Based on these dependent relationships, a complex network of dependencies among product components can be established. Then the components related to the source of change can be selected according to the complex network. At last, the change estimation can be carried out based on these relative components. The advantage of this estimation method is that fitting errors are used to resize the step size of the fitting, and it has good adaptability. Several groups of tests show that the proposed method not only solved the problem that multiple sources of change exist simultaneously, but also met the requirement error of the referenced paper.