The OPTHiNOS connects different sophisticated numerical approaches for Sensitivity Analysis (SA), Uncertainty Quantification (UQ), design and optimization of complex systems under chaos condition.

Leveraging from statistical science and scientific computing, the OPTHiNOS is the first proven approach to solve high-fidelity, large-scale and non-linear optimization problems.

The OPTHiNOS is applied to both convex and non-convex optimization problems. Literally, this package leverages numerically efficient algorithms to improve the optimization progress of high-fidelity (large-scale PDE) problems.

The OPTHiNOS can be applied for a various range of designs, where the turbulent flow, chaos, strong non-linearity, and unsteadiness is matter.

Unlike conventional methods in the ROM-based optimizations, the OPTHiNOS relies on a new strategy to reduce the computational costs. Instead of training the closure models with multiple datasets, the OPTHiNOS uses the least amount of data needed for the training process. However, the OPTHiNOS applies different novel strategies for the training progress.