SmartUQ

AI and ML

Physics AI, Machine Learning & Surrogate Modeling

Physics-based simulations are used to train advanced Gaussian process-based surrogate models to accurately predict a wide variety of results, requiring fewer computationally heavy simulation runs or expensive tests relative to deep learning methods.

Data and analytics

Data Sampling, Subsampling & Design of Experiments

Intelligent space-filling and uniform subsampling algorithms compress massive simulation and test datasets, preserving critical feature distributions and parameter sensitivities while slashing computational overhead. Novel DOEs and adaptive sampling methods maximize coverage with minimal simulation or test runs.

AI and ML

Uncertainty Quantification & Sensitivity Analysis

Comprehensive global sensitivity analysis, aleatory/epistemic uncertainty propagation, and reliability-based optimization ensure robust, risk-informed engineering decisions.

AI and ML

Statistical Calibration & Model Validation

Advanced Bayesian calibration systematically tunes model parameters against physical test data accounting for measurement noise and discrepancy. Model validation assesses agreement between simulations and physical data aligned with ASME V&V, FDA, DoD, and NASA standards.

AI and ML

Digital Twins & Predictive Analytics

Builds fast, calibrated virtual sensors and powers real-time digital twins for condition-based maintenance (CBM+), prognostics, and what-if analysis, enabling live health monitoring and downtime avoidance.

Advisory and strategy

Consulting Services

SmartUQ's team of engineers and data scientists helps customers assess requirements, develop customized analytics tools, integrate SmartUQ into engineering workflows, and produce decision-quality results from existing data sets via proof-of-concept projects.

Interactive graphical environment for data processing, visualization, creation of DOEs and emulators, integration and automation of other software, and performing analytics including uncertainty propagation, sensitivity analysis, inverse analysis, statistical calibration, and statistical optimization.

Our assessment: Flagship productized GUI platform with a broad, well-described feature set central to the firm's offering.

Extensive Python application programming interface with similar features to the SmartUQ GUI, used for automation, scripting, complex workflow creation, and interfacing with other software to fit emulators and run analytics.

Our assessment: Productized Python API tightly integrated with the SmartUQ platform, supporting advanced automation and workflow use cases.

Python API enabling users to programmatically use previously trained SmartUQ emulators to make predictions, designed to distribute emulators within other workflows and software.

Our assessment: Branded deployment-focused API for embedding trained emulators; described but with limited detail beyond its distribution purpose.

Package combining the SmartUQ GUI, PySmartUQ API, and Runtime API for easy access to SmartUQ features within Python.

Our assessment: Branded bundled package; substantive insofar as it consolidates core products, but described only briefly.

Adds the Prediction GUI and Analytics GUI to SmartUQ Pro, providing enhanced interactive environments for predictions and analytics workflows.

Our assessment: Tier-above bundle with added GUIs; named and described but no deeper technical differentiation stated.

Add-on module that allows users to build advanced emulators for complex varying geometry problems, including spatial-temporal emulators of simulations with inconsistent mesh or topology or from physical sampling across non-uniform points.

Our assessment: Specialized add-on for a niche but technically distinct engineering use case; has its own linked page suggesting moderate depth.