Reliability analysis for manufacturing system of drive shaft based on dynamic Bayesian network

Accurately analyzing the reliability of driveshaft systems is crucial in engineering vehicles and mechanical equipment. A complex system reliability modeling and analysis method based on a dynamic Bayesian network (DBN) is proposed to repair accurately and reduce the cost in time. Considering the logical structure of the drive shaft system, the reliability block diagram (RBD) of the manufacturing system is constructed in a hierarchical and graded manner, and a method of obtaining the Bayesian network (BN) directly from the RBD is adopted based on the conversion relationship between the RBD, fault tree and BN. A variable-structure DBN model of the system is constructed based on a static BN extended in time series and incorporating dynamic reliability parameters of the components. Reliability analyses based on DBN reasoning, including reliability assessment, significance metrics, and sensitivity analyses, were performed to identify critical subsystems and critical components. This research contributes to enhancing product reliability, equipment utilization, and improving economic efficiency.