A NOA-AP based Typical Process Route Automatic Discovery Approach for Intelligent Manufacturing

With the continuous advancement of intelligent manufacturing, the efficient extraction of reusable typical process routes based on artificial intelligence for computer-aided process planning (CAPP) has emerged as a critical necessity. Because a good process route can not only effectively improve the efficiency of process planning, but also promote the transformation of traditional manufacturing to intelligent manufacturing. To address this situation, this paper proposes a Nutcracker Optimization Algorithm (NOA) enhanced Affinity Propagation (AP) algorithm under the consideration of clustering effectiveness to automatic discover the typical process routes from the massive historical data. First, two soft constraints (i.e., quantity constraint and size constraint) are introduced simultaneously in the clustering process to ensure the quality and validity of the obtained typical process routes according to the actual demand. In addition, previous research has emphasized that the selection of preference value and damping factor in the AP algorithm significantly influences clustering outcomes. Hence, NOA is implemented to identify the optimal individual within the search space to determine suitable preference values and damping factors, thereby enhancing the clustering quality of the AP algorithm. The numerical illustration shows that the proposed NOA-AP could improve the clustering quality and discover more representative and effective typical process routes.