Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{201899,
        author = {Vinay Hanmant Dharme and Prof. P. S. Ladgaonkar},
        title = {Automatic Generation of Product Manufacturing Information From 3D CAD Models Using Deep Learning: Experimental Validation and Post-Result Analysis},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {4958-4962},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=201899},
        abstract = {Product Manufacturing Information (PMI) play a critical role in the Model-Based Definition (MBD) by embedding manufacturing semantics directly into 3D CAD models. Traditional PMI creation is primarily manual, requiring significant engineering effort and domain expertise. Although previous research has proposed deep learning–based frameworks for PMI prediction, limited work has demonstrated complete experimental validation and post-result performance evaluation in industrial CAD environments. This paper presents an enhanced deep learning–based framework for automatic PMI generation from 3D CAD models with experimental validation. The proposed method utilizes STEP AP242 CAD models, extracts Boundary Representation (B-Rep) information, converts geometric topology into graph structures, and applies a Graph Neural Network (GNN) to predict manufacturing annotations. Unlike previous conceptual approaches, this study evaluates the proposed model using actual CAD datasets and quantitative performance metrics. Experimental results show that the developed system successfully predicts dimensions, tolerances, datum references, and manufacturing annotations with high accuracy. The graph-based representation preserves geometric fidelity and improves prediction capability compared to voxel-based approaches. The developed framework demonstrates strong potential for reducing manual drafting effort and enabling intelligent PMI automation in digital manufacturing workflows.},
        keywords = {CAD Automation, CAD Intelligence, Graph Neural Network, Manufacturing Annotation, Model-Based Definition, Product Manufacturing Information, STEP AP242.},
        month = {May},
        }