Explainable artificial intelligence for manufacturing cost estimation and machining feature visualization

Studies on manufacturing cost prediction based deep learning have begun in recent years, but the rationale cannot be explained because models are still used as a black box. This study aims to propose process for 3D computer-aided design (CAD) using explainable artificial intelligence. The proposed can visualize machining features of CAD model that influencing increase costs. consists (1) data collection and pre-processing, (2) architecture exploration, (3) visualization explain results. shows high predictability computer numerical control (CNC) machined parts. In particular, gradient-weighted class activation mapping proves not only detect CNC also differentiate difficulty same feature. Using process, we provide guidance engineering designers reducing costs during conceptual phase. We real-time quotations redesign proposals online platform customers.