Abstract: During the product evolution design, the two key issues troubled by the designer are design data complex range and design rules difficult extraction. A reduction algorithm based on rough set and an extraction algorithm based on classification consistency rate, are presented. The former completes the product data reduction, the latter achieves the design knowledge mining and rules extraction. Results indicate the effectiveness and high efficiency of the algorithm.

Key words: rough set, product data reduction, design knowledge mining, design rules extraction, empirical analysis

摘要： 产品演进设计中，产品数据复杂繁多、设计规则难以提取是困扰设计人员的两个关键问题。提出基于粗糙集理论的属性约简算法，较好地解决了产品数据约简问题;利用分类一致性算法，较好地解决了产品设计知识挖掘和设计规则提取问题。通过实证分析，证明了算法的有效性和高效性。

关键词: 粗糙集理论, 产品数据约简, 设计知识挖掘, 设计规则提取, 实证分析