Empowering R&D with Data Analysis and Uncertainty Quantification

Uptimai is a high-tech company with a unique AI-powered software that together with an experienced engineering team, allows you to have deeper insights from your simulations and to optimize your products for real-life operation.
Empowering R&D with Data Analysis and Uncertainty Quantification

Uptimai is a high-tech company with a unique AI-powered software that together with an experienced engineering team, allows you to have deeper insights from your simulations and to optimize your products for real-life operation.

How it works

Your non-optimized design

Your products and designs are modern, but highly complex with many dependencies or lots of data types.

Coupling

We couple Uptimai software with your simulation platform (Nastran, Ansys, Abaqus…), with your dataset or API interfaces.

Preparation

The ranges for all your parameters are chosen to establish the domain to be analyzed.

Computing

The Uptimai Machine Learning algorithm builds an accurate model propagating all the uncertainties across all inputs.

Graphical results

From the created model and thanks to our post-processing tools, you can obtain deeper insights needed to optimize your design.

Case studies

Retainer Optimization

Retainer Optimization

Design, analysis and optimization of retainer materials

Centrifugal Fan

Centrifugal Fan

Turbomachinery CFD study

Car Box Beam Deformation

Car Box Beam Deformation

Absorbed impact energy increased by 10%

Testimonials

Virtual Vehicle Research GmbH
Dr. Matthias K. Scharrer
Battery research group, Virtual Vehicle Research GmbH
With Uptimai we conducted an experiment on validation and parametrisation in electro-chemical battery modelling. The challenge was to find at least one suitable parameter set to describe measurements using a “single particle” model. The Uptimai algorithm accelerated the entire process, as the subsequent analysis was carried out much faster using the increasingly improving metamodel. Uptimai gave us a comprehensive insight into the importance of the dynamic parameters, as well as regions of preference and of avoidance. This new knowledge will be integrated into the future parameter estimation and model calibration workflow.
Dr. Matthias K. Scharrer, Virtual Vehicle Research GmbH
Karel Beneš, MECAS-ESI s.r.o.
Jan Šplíchal, VUT Brno
Jan Macháček, Altran s.r.o
Jakub Cejpek, Idiada a.s.

Partners

logo
logo
logo
logo
logo
logo
logo
logo

Send us a message

Email must not be empty
Phone number must not be empty
Enquiry must not be empty