What question did this study set out to answer?

To develop a framework that redirects generative CAD from geometric similarity to functionality in engineering designs.

July 4, 2026

From geometry to function: towards context-aware generative AI for engineering design

Key Points

To develop a framework that redirects generative CAD from geometric similarity to functionality in engineering designs.
Introduces a pre-learning strategy using synthetically labeled datasets evaluated by FEA, CAM, and LCA.
Proposes a self-learning strategy that utilizes knowledge-based tools in a reinforcement learning feedback loop.
Demonstrates improved consideration of functionality in CAD designs compared to traditional methods.
Shows potential for enhanced manufacturability and sustainability outcomes in engineering applications.

Abstract

ABSTRACT: Current generative artificial intelligence for Computer-Aided Design (CAD) optimizes for geometric similarity, neglecting engineering criteria like functionality, manufacturability, and sustainability. This paper addresses this gap and proposes a conceptual framework to reorient generative CAD from replicating shapes to achieving function. We introduce two hybrid training strategies: a pre-learning approach using synthetically labeled datasets (evaluated via FEA, CAM, LCA) and a self-learning approach where GenAI uses these knowledge-based tools as a reinforcement feedback loop.

Demander à l'IA

Bookmark

View Full Paper