SPDM 2024 R2: Integrating AI into Engineering with SPDM
Machine learning is set to play a central role in engineering. The growing possibilities require structured training data and traceability of model usage. Transparency is a key element for decisions based on AI models. Ansys Minerva’s Simulation Process and Data Management (SPDM) system is perfect for organizing training data and models. This makes it possible to know at any time which data was used, when, and for what purpose.
Machine learning methods are perfectly suited to typical product development tasks: accelerating development stages, providing feasibility statements and possible variants in a few seconds, for example in applications aimed at non-specialists in simulation.
Model training must be based on existing knowledge within the company, and this knowledge is found in existing simulation projects. Simulations also play a crucial role in generating training data when no real test data is available.
Traceability is a key element.
The possibilities arising from this will increase exponentially. Therefore, it is all the more important to focus on traceability and organization. If machine learning is to be integrated into the development process as a tool, the training data must be structured and versioned. Over time, new discoveries will emerge and require retraining. Traceability must remain consistent.
The use of an AI model must also be traceable if decisions are made based on this model. It is essential to know when the AI model was trained, with what data, and what decision was made based on this model.
Structuring and Traceability through Simulation Data Management
The underlying training data, AI models, and the resulting applications are typical data of CAE (Computer-Aided Engineering). Therefore, it makes sense to organize them through a simulation process and a data management system:
- Ansys Minerva specializes in mapping CAE processes, such as those required for the automated generation of training data.
- As a simulation data management system, it provides a data model designed to manage the large number of training datasets and the associated data volume.
- Le cœur de l’application de la ML en ingénierie est la traçabilité des dépendances, quelles données ont été utilisées à quel moment et dans quel but.
Written by CADFEM Germany GmbH
Contact us or watch our CADFEM webinars on the new version for additional advice and answers to your questions.