【摘要】 在车身生产过程中 ,平时积累了大量的抽检数据用于质量监控 ,然而由于车身测量数据的抽样小 ,不同总成抽测频率也不完全相同等特点 ,无法直接被用于车身制造偏差源的诊断中的相关性分析 ,当前用相关性分析来确定故障源的方法存在需要大量附加测量实验。本文提出了基于贝叶斯理论的相关性分析方法 ,可利用原有测量数据实现车身制造质量故障的快速诊断 ,有效的降低人力、物力资源的耗用 ,产生大的经济效益更多还原

【Abstract】 Lots of auto-checking data which is used to quality control is accumulated for a long time However, it can’t be used to relativity analysis of body manufactory variation because of the small simple and difference checking-frequency It is need lots added checking tests before relativity analysis for looking for the reason of manufactory problem This paper discusses a new relativity analysis way based on Bayesian theory, which can quickly fault diagnose by auto-checking data It can fall the cost of manpower and material resources and produce large economic benefit更多还原