SE Validation Limited

Validation and Verification (V&V) of Models and Simulations

Introduction

Our M&S validation and verification (V&V) capabilities are based on processes to determine the degree to which models and simulations accurately represent the intended real-world in the context of the conceptual model of the problem. By conducting V&V we can increase the confidence in a product through documenting and communicating detection (and subsequent elimination) of errors.

Planning

V&V plans need to identify the demonstration cases that are required by the customer and to include the V&V arguments and evidence that will be used to satisfy claims of verification or validation. We have experience in creating these cases such that the validation and verification of an item becomes part of the development cycle, and can be confirmed in an appropriate manner. Selection of particular V&V techniques is based on availability of information and referents and the risks associated with intended use of the M&S.

Implementing

Verification assesses how a product is constructed in the context of the original concept of intended use. There are many static and dynamic testing methods available and we have used or reviewed the results of many of them, including Control Flow Analysis, Data-use Analysis and Information Flow Analysis. We also have experience of formal mathematical proofing of algorithms. Validation is often based on the comparison between a product of interest and an appropriate representation of the real-world. Different levels of validation are applicable to different tasks, ranging from high level validation using real-world referents, trials and testing, through to lower level validation against theoretical evaluation and expert opinion.

Recording

Of course, all this information and evidence has to be recorded, collated and presented in a concise and accurate way to support critical decisions. Evidence needs to be assessed and appropriate cases for verification and validation need to be developed. With many years of experience in this field the staff at SE Validation are expert in this field and have represented UK in key international standardisation efforts.

Example Papers

Lies, Damn Lies and Computer Simulation Proceedings of ESM2000 Conference pp 72-76, Ghent 2000