Manufacturing Simulation Acceleration for Plastic Part Surface Field Prediction Using Altair PhysicsAI Machine Learning Model

Presentation by Axel Ramm, Amit Mandal, and Kirtesh Pawar with Whirlpool Corporation as part of the 2025 ATCx AI for Engineers conference.

Plastic components such as endcaps in refrigeration systems require extensive simulation cycles to optimize design and manufacturing parameters. Conventional CAE workfl ows—analyzing warpage, shrinkage, sink marks, pressure, clamp force, and cycle time—are often time-consuming and computationally intensive. To address this, we present a machine learning-based predictive framework powered by Altair PhysicsAI, aimed at significantly accelerating surface field result prediction for plastic parts.

The project was structured in five key phases. The primary objective was to develop a PhysicsAI-driven model capable of accurately predicting critical simulation outcomes for plastic endcaps.