What question did this study set out to answer?

This work aims to enhance predictive modelling in CAD by utilizing graph-based methods to predict design operations.

July 4, 2026

AI-enhanced computer-aided design: predictive modelling of operations

Key Points

This work aims to enhance predictive modelling in CAD by utilizing graph-based methods to predict design operations.
Converted real CATIA V5 models into directed acyclic graphs to capture feature dependencies.
Utilized a four-layer Graph Attention Network for predictions.
Compared performance against a frequency-based non-parametric baseline.
Achieved a top-5 prediction accuracy of 94%.
Outperformed frequency-based baseline significantly.
Validated that graph representations and attention mechanisms are effective for modelling assistance.

Abstract

ABSTRACT: This work introduces a graph-based CAD assistant that predicts the next modelling operation in parametric design sequences. Real CATIA V5 models from the automotive domain are converted into directed acyclic graphs capturing feature dependencies, enabling learning directly from structural design data. A four-layer Graph Attention Network achieved a top-5 prediction accuracy of 94%, outperforming a frequency-based non-parametric baseline. The results show that graph representations and attention-based message passing provide a strong foundation for context-aware modelling assistance.

Ask AI

Helpful

Bookmark