Hybrid Simulations

The increasing use of virtual simulation, from product design and manufacturing to field operations, makes it vital to manage its associated compute, storage, and service costs. The Hybrid Simulation solutions on AWS use artificial intelligence and machine learning (AI/ML) as well as high performance computing (HPC) data to build and train surrogate models for virtual simulation. A surrogate model can be a data-driven model trained to approximate the behavior of an original model. To generate fast responses for new designs, these solutions allow you to query a surrogate model, rather than running full-scale simulations. Additionally, by leveraging AI/ML capabilities, such as intelligent auto-steering of simulations guided by optimizers, it is possible to improve design analysis and mitigate operational issues. You can use these solutions to build higher order predictive models and perform both sensitivity analysis and uncertainty quantification to tackle even the most complex industrial designs.