a b s t r a c tBefore the results of a scientific computer simulation are used for any purpose, it should be determined if those results can be trusted. Answering that question of trust is the domain of scientific computer simulation review. There is limited literature that focuses on simulation review, and most is specific to the review of a particular type of simulation. This work is intended to provide a foundation for a common understanding of simulation review. This is accomplished through three contributions. First, scientific computer simulation review is formally defined. This definition identifies the scope of simulation review and provides the boundaries of the review process. Second, maturity assessment theory is developed. This development clarifies the concepts of maturity criteria, maturity assessment sets, and maturity assessment frameworks, which are essential for performing simulation review. Finally, simulation review is described as the application of a maturity assessment framework. This is illustrated through evaluating a simulation review performed by the U.S. Nuclear Regulatory Commission. In making these contributions, this work provides a means for a more objective assessment of a simulation's trustworthiness and takes the next step in establishing scientific computer simulation review as its own field.Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

No abstract

No abstract

This paper provides a mathematically complete set of errors for modeling and simulation. First, a general scenario is introduced which describes a set of relations (e.g., functions) and inputs which are common to many, if not most, applications of modeling and simulation. Second, using those relations and functions, an equation for total error is given. That equation is manipulated by introducing important terms from the general scenario to obtain a set of errors which are mathematically equivalent to the total error. By deriving the set of errors in this manner, the set of errors must be a complete set. That is, there cannot be an error in the model or simulation which is not captured by one of the errors introduced. Finally, each individual error is discussed to understand what that term is assessing and how that term is typically addressed in uncertainty quantification.

No abstract