Skip to content
Contact us

Machine learning for material and process optimization

A new AI based tool to reduce experimental time and optimize process parameters

Easy to use, Fast, Accurate, AI

Unique Deep infinite mixture of Gaussian Processes (DIM-GP) method. Combines the benefits of Neural networks and Gaussian process methods in a single AI/ML model designed to work with small datasets.

Paint optimization design of experiment

AI designed for material and process optimization

Reduce the number of experiments to get the right material and process parameters

  • Build AI from a small number of experiments
  • Probabilistic machine learning recommend experimental parameters based on goals

Reusable AI model

  • Predict impact of material and process changes

AI Assisted Design Of Experiements (DOE)

  • Minimize number of experiments (physical or numerical)
  • AI based result analysis
  • AI based experimental recommendations
AI Assisted DOE

Solving a key challenge of AI models

  • High accuracy
  • From minimal data (10 to 100 data points)

AI powered adaptive DOE

Probabilistic model with uncertainly quantification

Accepts various types of data

Scalars, signals, fields, tensors, images, meshes can be used as input / output:

Would you like to reduce the number of long and costly experiments?

Fil in the form below

Learn more

Example

Paint composition and process optimization

 

formulation example
Input parameters
output parameters constraints
AI experimental results
AI experimental results visual
prediction composition

Machine learning for efficient material and process

Our solutions help you:

  • Search for the optimal formulation with few experiments

  • Creation of digital twins of your products and processes

  • Optimization of process parameters

  • Identifying the most important parameters for your product or process

  • Optimal choice of raw materials and the replacement of existing raw materials with substitutes