AIMold: An Autonomous AI-based Pipeline for Complex Mold Design

2026-08-01

AI Summary

This study addresses the longstanding reliance on expert knowledge in designing injection molds for complex plastic parts, a process hindered by the absence of automated methods and publicly available data. To bridge this gap, the authors introduce MoldCAD, the first large-scale structured CAD dataset dedicated to complex mold design, comprising 4,934 parts and 3,850 complete mold assemblies. They further propose an end-to-end AI pipeline that leverages deep learning to predict parting directions, generate parting surfaces, identify auxiliary components, and perform assembly reasoning—directly producing manufacturing-ready mold designs. The method successfully generates over 23,000 fabricable mold components, achieving, for the first time, fully automatic translation from part geometry to complete mold assembly and advancing the frontier of manufacturing-aware CAD generation.

Abstract

Injection molding is the cornerstone of mass-producing plastic components. While current algorithms can automate mold design for basic geometries using standard two-piece molds, complex parts featuring undercuts, side holes, or re-entrant features present a significant challenge. These geometries often necessitate auxiliary components beyond the primary upper and lower molds. In practice, designing these intricate assemblies is a laborious process that relies heavily on expert knowledge. Furthermore, the scarcity of public datasets has hindered the development of effective learning-based solutions. To bridge these gaps, we introduce MoldCAD, a curated dataset that pairs complex single-body CAD parts with industry-standard mold assemblies. Each entry includes the upper and lower molds, parting surfaces, demolding orientations, and necessary auxiliary components. The dataset comprises 4,934 CAD models and over 3,850 mold assemblies, totaling more than 23k individual models. Building upon this dataset, we propose a comprehensive pipeline that predicts demolding orientations, identifies auxiliary components, and constructs parting surfaces to derive a complete, manufacturing-ready mold assembly for downstream CAD/CAM workflows. Our results demonstrate a promising path toward fully automated industrial mold design and contribute to the broader advancement of manufacturing-aware CAD generation.

Problem

• injection molding
• complex geometries
• mold design automation
• auxiliary components
• CAD dataset

Innovation

• autonomous mold design
• MoldCAD dataset
• demolding orientation prediction
• parting surface generation
• auxiliary component identification