Collectively the processes involved in evaluating our level of trust in the results obtained from models are known as VVUQ. While the need for rigorous model assessment is widely acknowledged, it is far from being universally implemented within the scientific literature. The reasons for this are wide ranging, but include lack of specialist knowledge of VVUQ techniques and, until recently, the difficulty in obtaining sufficient computational power to perform the necessary sampling in large scale

more »

... simulations. We have recently developed EasyVVUQ, [3] a package designed to help leverage recent advances in the scale of computational resources to make state of the art VVUQ algorithms available and accessible to a wide range of computational scientists. EasyVVUQ is a component of the VECMA open source toolkit (http://www.vecma-toolkit.eu), which provides tools to facilitate the use of VVUQ techniques in multiscale, multiphysics applications. [4] n order to enable straightforward computations of EasyVVUQ scenarios on HPC resources, the tool has been designed to work with a variety of middleware technologies, such as FabSim3 [5] or