Evaluation of a design property network-based change propagation routing approach for mechanical product development

Design changes are unavoidable in new mechanical product development, and the propagation of changes imposes negative impacts on the design cycle and cost. Due to the non-uniqueness of changing propagation paths, searching for the optimal change propagation path with minimum change-related impact tends to be a serious challenge before the implementation of the change. In this paper, a mathematical programming model is presented to route change propagation. First, the design change analysis model (DCAM) is built based on the design property network. In the DCAM, the design properties are connected by linkages, and the weights of these linkages are objectively estimated by mining the change records. In addition, the change propagation intensity (CPI) is defined by quantifying the change propagation impact. The CPI is indicated by four assessment factors: propagation likelihood, node degree, long-chain linkage, and design margin. Furthermore, the optimal influence propagation path, which corresponds to the minimized maximum of accumulated CPI, is sought with a modified ant colony optimization (ACO) algorithm. A case study and solution comparison verify the feasibility and validity of the proposed approach.