Leveraging Vision-Language Models for Manufacturing Feature Recognition in Computer-Aided Designs

Abstract

Automatic feature recognition (AFR) is essential for transforming design knowledge into actionable manufacturing information. Traditional AFR methods, which rely on predefined geometric rules and large datasets, are often time-consuming and lack generalizability across various manufacturing features. To address these challenges, this study investigates vision-language models (VLMs) for automating the recognition of a wide range of manufacturing features in computer-aided designs (CAD) without extensive training datasets or predefined rules. Instead, prompt engineering techniques, such as multiview query images, few-shot learning, sequential reasoning, and chain-of-thought, are applied to enable recognition. The approach is evaluated on the proposed CAD dataset containing designs of varying complexity relevant to machining, additive manufacturing, sheet metal forming, molding, and casting. Five VLMs, including three closed-source models (GPT-4o, Claude-3.5-Sonnet, and Claude-3.0-Opus) and two open-source models (LLava and MiniCPM), are evaluated on this dataset with ground truth features labeled by experts. Key metrics include feature quantity accuracy, feature name matching accuracy, hallucination rate, and mean absolute error (MAE). Results show that Claude-3.5-Sonnet achieves the highest feature quantity accuracy (74%) and name matching accuracy (75%) with the lowest MAE (3.2), while GPT-4o records the lowest hallucination rate (8%). In contrast, open-source models have higher hallucination rates (larger than 30%) and lower accuracies (less than 40%). This study demonstrates the potential of VLMs to automate feature recognition in CAD within diverse manufacturing scenarios.