Product design to the next level

Materialize offers a material design platform, enhanced by a multiphysics simulation engine, harnessing the power of the cloud!

A high-end material design tool

Materialize offers a vast database coupled with a simulation engine, ready to provide material selection choices with a click of a button!

High-end modeling

Materialize is a high-end modelling software for the analysis and design of high performance materials. It is highly scalable and parallel with the ability to perform simulations of complex real-world structures comprised of the material under design, providing the designer with a stochastic multiscale optimization framework which integrates multiple lengths and time scales.

Material response optimization

Materialize provides the ability to calculate and optimize the material response, either on its own or as part of a real-world complex structure, while at the same time, simulate and consider the effects of randomness of both the material structure and the properties of the structure comprised of this material. Materialize addresses the needs of several industrial sectors (chemical, building construction, electronics, automotive, aerospace and energy industries) whose needs are not met by the current off-the-shelf software.

A coherent stochastic multiscale optimization framework

Materialize is an innovative robust stochastic multiscale optimization (SMO) computational platform. The multiscale problem is formulated in a hybrid hierarchical/concurrent FE2 setting as a nested solution scheme between a macroscale finite element (FE) model and realistic representative volume elements (RVEs) characterizing the material points of the structure. A hierarchical approach constructs the corresponding RVE FE models at the micro-meso scale, starting from the description of carbon inclusions at the atomic level and upscaling to FE representations of realistic RVEs, considering uncertainties involved in the microstructural topology. Interphase properties at the molecular level are derived via detailed molecular dynamics (MD) and/or molecular mechanics (MM) models and projected to efficient surrogate models that are integrated in the SMO platform.

HPC ready

At its core, Materialize implements highly scalable parallel algorithms for solving the extremely large-scale systems of algebraic equations, as they occur for the solution of SMO problems. This high demand for computing resources is addressed with domain decomposition methods (DDMs), custom tailored to the properties of a SMO problem. Materialize features implementations of these algorithms in hybrid CPU/GPU computing environments in order to minimize computing time and at the same time, utilize the heterogeneous resources of contemporary HPC environments.