Typical process discovery of satellite typical parts based on affinity propagation clustering

Aiming at the problem that a lot of repeated jobs existed in the process planning of satellite typical parts, and many kinds of historical process data were not fully reused, the typical process discovery was researched to improve the utilization of typical process and enhance the process planning efficiency. The discovery problem of typical process was described, and the attributed directed graph model of process was built. The similarity between processes was obtained with the similarity measurement of process cells and process routes. Based on the process similarity, affinity propagation method was used to cluster the process, and the clustering result was evaluated according to Silhouette index and In-Group Proportion (IGP) index to get the best clustering number. The machining process of satellite plate was used as an example to verify the effectiveness of proposed method, and five typical processes were obtained. ©, 2015, CIMS. All right reserved.