Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
August 24, 2026Journal of Intelligent ManufacturingOpen Access

A hybrid machine learning framework for robust feasibility prediction of 3D mechanical designs using scalar and geometric features

View Full Paper
Ask AI
Bookmark
Share

Authors

MFMd Mohsin Uddin FahimALAmit J. Lopes

Discussion

Loading...

Member takes

Overview

Validation study demonstrates accurate manufacturing feasibility prediction in 3D mechanical designs, indicating effective early-stage screening through hybrid feature modeling.

Key Points

  • To develop and evaluate a machine learning framework that combines scalar design parameters and 3D geometric features to predict the manufacturing feasibility of mechanical designs at early development stages.
  • Analyzed an industrial dataset of 654 mechanical design instances labeled by real-world manufacturing outcomes, evenly balanced between feasible and infeasible cases.
  • Combined standardized scalar parameters with 3D geometric descriptors (such as volume, aspect ratio, and bounding box dimensions) through statistical analysis and feature selection.
  • Trained and evaluated multiple classifiers using stratified cross-validation and a group-wise protocol that withheld whole design groups to assess generalizability.
  • The framework achieved approximately 81% internal accuracy and around 84% accuracy on unseen design groups, with a ROC-AUC of approximately 0.89.
  • Ablation analysis revealed that scalar constraints provided the majority of predictive signal, while geometric descriptors provided marginal incremental improvement.
  • A three-zone decision policy was established using calibrated probabilities to separate automatic acceptance or rejection from borderline designs requiring manual review.

Cite This Study

Fahim et al. (2026) studied this question.

synapsesocial.com/papers/6a8c00d0bca056c88e6dfc36https://doi.org/10.1007/s10845-026-02962-9
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Machine Learning-Based Manufacturing Cost Prediction from 2D Engineering Drawings via Geometric Features2025
  2. 2A Novel Deep Hybrid Learning Framework for Structural Reliability Under Civil and Mechanical Constraints2025 · 10 citations
  3. 3A decision support framework for garment fit prediction: standardizing fabric mechanical properties in 3D virtual prototyping2026
  4. 4Automated parametric design framework for patient-customized wearable devices2026
  5. 5A Novel Machine Learning Framework for Digital Estimation of Allowables in Aerospace Composites2024 · 3 citations