A method for discovering typical process sequence using granular computing and similarity algorithm based on part features

Aiming at the problems of hereditability and reusability for process planning data in the enterprises, a novel method for discovering typical process sequence by applying granular computing and similarity algorithm based on part features is put forward. Firstly, a calculating model of process sequence similarity is built through the analysis and comparison of part features, including machining features and topological relations between features, and a fuzzy similarity matrix of all the process sequences is established. Afterwards, according to the theory of fuzzy quotient space, which is one of the theoretical models of granular computing, a process sequence quotient space family with hierarchical structure is constructed. The granularity of every process sequence quotient space is measured by information entropy, and the information gain, which means the difference of information entropy between two adjacent quotient spaces, is calculated. Finally, a quotient space with the bigger information gain as well as the higher minimal process sequence similarity in process information granules is determined as an optimal process information granular layer, in which some typical process sequences are acquired from process information granules by using longest common subsequence algorithm. An application example verifies the feasibility and validity of the proposed method.