Uncertainty quantification (UQ) methods are essential for designers, engineers, and scientists to make precise statements, as well as quantify numerically, the degree of confidence they have in their simulation-based decisions.
Quantifying the total uncertainty in a simulation will ensure decision-makers can measure the credibility of a prediction and proactively work to save time, allocate resources, and reduce the risk of inadequate safety, reliability, or performance of the system. Everyone has a model but it is challenging to communicate the approximations, assumptions, and uncertainties that exist in any model prediction. Utilizing a UQ framework will provide you with a consistent and proven way of using the uncertainty in model predictions to make risk informed decisions.
This course explains the concepts and effective procedures used not only to predict uncertainties in a model, but to also mature your model and build trust in your organization by being able to communicate and document your findings. This systematic framework focuses on methods, approaches, and strategies for quantifying uncertainties in model predictions.
Apply what you learn! Probabilistic and UQ methods are presented in-depth followed by exercises to reinforce the material. Attendees will learn how to use the NESSUS probabilistic analysis software and will apply it throughout the course to gain experience in problem formulation and results interpretation and communication.
By participating in this course, you will learn how to successfully:
Identify potential uncertainties in models and data.
Represent uncertainties in models and inputs.
Explain how uncertainties impact model predictions.
Select methods to efficiently propagate uncertainties in the models.
Identify options to reduce uncertainties in the model predictions.
Who should attend?
This course is essential for engineers, scientists, and technical managers concerned with managing uncertainties in model predictions used to make decisions in the engineering design and evaluation process.
Course Materials (included in purchase of course)
Digital course notes via ASME’s Learning Platform
Software access to NESSUS for 90 days Attendees will need a Windows computer to complete the course exercises. Download and software installation instructions will be provided prior to the course and can be installed prior or during the course.
Required Course Materials (not included with course, purchase separately)
Attendees will need a Windows based laptop computer to complete the course exercises.
Attendees must have administrator permissions in order to install the software.
Topics covered in this course include:
Modeling uncertain variables
Propagating uncertainties
Formulating Uncertainty Qualification (UQ) problems
Sensitivity analysis
UQ for numerical models
Response surface models for efficient uncertainty propagation
Bayesian statistics for uncertainty quantification
Model Parameter Calibration
UQ solution strategy examples and case studies
This ASME Virtual Classroom course is held live with an instructor on our online learning platform. Certificates of completion will be issued to registrants who successfully attend and complete the course.
Uncertainty quantification (UQ) methods are essential for designers, engineers, and scientists to make precise statements, as well as quantify numerically, the degree of confidence they have in their simulation-based decisions.
Quantifying the total uncertainty in a simulation will ensure decision-makers can measure the credibility of a prediction and proactively work to save time, allocate resources, and reduce the risk of inadequate safety, reliability, or performance of the system. Everyone has a model but it is challenging to communicate the approximations, assumptions, and uncertainties that exist in any model prediction. Utilizing a UQ framework will provide you with a consistent and proven way of using the uncertainty in model predictions to make risk informed decisions.
This course explains the concepts and effective procedures used not only to predict uncertainties in a model, but to also mature your model and build trust in your organization by being able to communicate and document your findings. This systematic framework focuses on methods, approaches, and strategies for quantifying uncertainties in model predictions.
Apply what you learn! Probabilistic and UQ methods are presented in-depth followed by exercises to reinforce the material. Attendees will learn how to use the NESSUS probabilistic analysis software and will apply it throughout the course to gain experience in problem formulation and results interpretation and communication.
By participating in this course, you will learn how to successfully:
- Identify potential uncertainties in models and data.
- Represent uncertainties in models and inputs.
- Explain how uncertainties impact model predictions.
- Select methods to efficiently propagate uncertainties in the models.
- Identify options to reduce uncertainties in the model predictions.
Who should attend?
This course is essential for engineers, scientists, and technical managers concerned with managing uncertainties in model predictions used to make decisions in the engineering design and evaluation process.
Course Materials (included in purchase of course)
- Digital course notes via ASME’s Learning Platform
- Software access to NESSUS for 90 days Attendees will need a Windows computer to complete the course exercises. Download and software installation instructions will be provided prior to the course and can be installed prior or during the course.
Required Course Materials (not included with course, purchase separately)
- Attendees will need a Windows based laptop computer to complete the course exercises.
- Attendees must have administrator permissions in order to install the software.
Topics covered in this course include:
- Modeling uncertain variables
- Propagating uncertainties
- Formulating Uncertainty Qualification (UQ) problems
- Sensitivity analysis
- UQ for numerical models
- Response surface models for efficient uncertainty propagation
- Bayesian statistics for uncertainty quantification
- Model Parameter Calibration
- UQ solution strategy examples and case studies
This ASME Virtual Classroom course is held live with an instructor on our online learning platform. Certificates of completion will be issued to registrants who successfully attend and complete the course.
