中文摘要：

复杂质量结构产品系统中质量要素多且相互关系复杂，在对复杂质量结构产品进行界定的基础上，以贝叶斯网络表征复杂质量结构产品的质量相关关系，依据贝叶斯推理，有效利用经验数据及先验概率，得到贝叶斯网络中对根节点影响大的关键节点，即得到复杂质量系统中的关键质量要素，为复杂质量系统建模提供了一种新的思路。最后以汽车发动机的贝叶斯网络建模分析为例，验证了文中所提方法的可行性。

英文摘要：

The elements of products with complex quality structure are of great numbers , and the relationships between them are also quite complex .Based on the definition for products with complex quality structure , Bayes-ian network is applied to express the correlativity in quality for such products .According to Bayesian reasoning , the key nodes which affect the root node significantly are figured out with the effective use of empirical data and prior probability .In this way , the key elements of the complex quality system are thus figured out , which also provides a new train of thought for complex quality system′s modelling and analysis .The example of car engine′s modeling and analysis process based on Bayesian network is finally given to verify the feasibility of the proposed method.