Packaging Drop Testing, Monolith AI

Use cases


Optimise the shape and position of a spoiler to optimise race car performance.
Get Started


Optimise the shape of a wind turbine to maximise performance while ensuring structural integrity.
Get Started

How it works

Get your data

Monolith uses your historical data from previous design cycles to learn. Export your simulation data from your preferred software, be it FEA or CFD, and load it directly into Monolith. You can work with simulation data in either tabular format or the raw 3D itself.



Previously, parameterization relied on being able to edit and tune tabular information. But what if your geometry is too complex to be described this way?

Our advanced autoencoders technology convert and compress your 3D designs into a set of parameters, allowing the defining characteristics of the product geometry to be identified and quantified.

Performance prediction

You can build predictive machine learning models and train them on your existing performance data. These are then capable of telling you the outcome of a new simulation instantaneously, including predicting new 3D fields of data, facilitating data-driven decision making between teams.



For a known target performance, the autoencoder can provide you the parameters for the performance-optimised 3D design and even rebuild your geometry in a ready to export format, reducing the number of simulations required to achieve peak performance.