Abstract: A rough set-based case retrieval technology was presented in the conceptual design of Flexible Manufacturing Systems (FMS). Design cases were described as a decision table that was formed by demand features and configuration features of systems. Rough set theory was applied to calculate the important degree of each feature attribute and remove the redundant ones, then the most important features were selected to retrieve the most similar case in case database. And to deal with the quantitative features, a method using grade classification and decision attributes support degree were put forward. The proposed method was also demonstrated by an application example .The results objectively revealed the similarity degree between the new problem and similar case, and the veracity and efficiency of case retrieval has been improved.

Key words: flexible manufacturing system, conceptual design, rough set, case retrieval

摘要： 针对柔性制造系统概念设计中的实例检索问题，提出了一种基于粗糙集理论的实例检索方法。分析了系统的需求特征和结构特征，通过建立属性决策表描述了实例信息；利用粗糙集理论对属性进行了重要性的判断和约简。为了处理属性表中的定量问题，给出了一种基于分级聚类法和决策属性支持度的离散化方法；抽取重要的特征属性作为检索依据，利用相似学方法进行相似度量，检索出最为接近的设计问题的实例作为设计参考。通过应用实例，给出了这种方法的具体实现过程。结果表明，该方法能够客观地反映实例与待求解问题的相似程度，保证了检索的正确性，提高了检索效率。

关键词: 柔性制造系统, 概念设计, 粗糙集理论