Root cause tracing of helicopter quality issues based on bayesian networks

In response to the challenges faced in assembly quality control during the intelligent transformation of the aerospace manufacturing industry-such as high data complexity, difficulties in root-cause tracing of quality issues, and the lag of traditional management approaches-this study investigates a Bayesian network-based method for causal analysis and root-cause tracing of helicopter assembly quality problems. By integrating the logical relationships between assembly processes and quality characteristics, a quality-characteristic Bayesian network model is developed using structure learning and parameter learning techniques, and probabilistic inference is applied to trace the causes of observed quality issues. The results demonstrate that this approach effectively captures the complex dependencies among multi-source factors in the assembly process, accurately identifies the key causes of quality problems, and provides a feasible technical pathway and methodological support for quality diagnosis, risk identification, and control optimization in helicopter assembly.