Identifying Key Parts of Complex Mechanical Products Based on Evidence Theory

The identification of key parts is of great significance in the whole life cycle of complex mechanical products. In this paper, a new method to measure the importance of parts is proposed. Firstly, a weighted parts association network (WPAN) is developed, taking into consideration assembly relationships and significant measurement relationships. Secondly, strength centrality and closeness centrality are integrated based on Dempster-Shafer evidence theory to identify key parts in WPAN. Then the susceptible-infected-susceptible (SIS) model is used for the evaluation of the method. Finally, an industrial case is used to verify the feasibility and effectiveness of the proposed method.