Model Validation & Uncertainty Quantification



Zhu Mao
University of Massachusetts Lowell

Roland Platz
Fraunhofer Institute for Structural Durability and System Reliability LBF

Garrison Flynn
Los Alamos National Laboratory





The purpose of the Technical Division on Model Validation and Uncertainty Quantification (MV&UQ) is to provide maximum technical benefits to its members by:


  • Promoting the advancement and understanding of MV&UQ theory and methods by providing a forum for the exchange of information on MV&UQ theory, methods, applications, and related activities
  • Promoting best practice in the areas of experimental testing and validated simulation through the use of MV&UQ methods
  • Providing educational opportunities to promote the understanding and use of proven MV&UQ methods
  • Whenever possible, promoting the exchange of information and coordinating MV&UQ related activities with other technical or professional societies


Best Paper Award Guidelines

Click here to download guidelines


SEM Symposium on Model Validation and Uncertainty

The first SEM Symposium on Model Validation and Uncertainty Quantification will be held at IMAC XXXV. The symposium has grown out of the sustained interest in MV&UQ activities at the IMAC conference, and is intended to provide a forum for the exchange of ideas in the development and application of MV&UQ techniques. A primary goal of the symposium is for participants to collaborate in the development of methods to address real-life structural dynamics applications. The symposium will include a keynote talk by Prof. Isaac Elishakoff (Distinguished Research Professor, Florida Atlantic University).


Numerical models and simulations are approximate representations of the actual systems they represent. Verification and Validation (V&V) along with uncertainty quantification (UQ) activities provide a means to establish model credibility in a quantitative and objective manner. Model verification ensures that the mathematical model is being solved correctly while model validation ensures that model is sufficiently credible representation of reality. Uncertainty quantification seeks to evaluate the effects of uncertainties that originate from numerous sources and track the propagation of those uncertainties to the final prediction(s) of the model. MV&UQ processes are necessary to ensure that advanced numerical models may be relied upon with confidence.

The symposium is intended to be of interest to both newcomers and experienced hands in this field, and should appeal to all conference attendees whose work makes use of model predictions. The highlight of the Monday sessions will be a keynote talk by Prof. Elishakoff followed by an overview of MV&UQ approaches presented by experts in the field. A full programme of talks over the following three days will cover a broad span of recent developments, from the development of new tools through to industrial application. Highlights include special sessions on modelling of musical instruments and UQ for material models, plus a session on novel and potentially powerful methods for controlling uncertainty. The first symposium will include sessions covering:

  • Overview of Model Validation and Uncertainty Quantification (2 sessions)
  • Uncertainty Quantification in Structural Dynamics (5 sessions)
  • Model Validation and Uncertainty Quantification: Industrial Applications
  • Controlling Uncertainty
  • Uncertainty Quantification in Material Models (2 sessions)
  • Modelling of Musical Instruments
  • Uncertainty in Early Stage Design
  • Applications of Uncertainty Quantification
  • Additive Manufacturing